My AI Note-taking workflows - CourtneyR.dev's 2nd Brain

My AI Note-taking workflows ![rw-book-cover](https://readwise-assets.s3.amazonaws.com/static/images/article2.74d541386bbf.png) 🔗 **Source:** [My AI Note-taking workflows](https://www.youtube.com/watch?v=2Jl6H9Ri8iI) > [!note]+ Document Notes > tags: [ai-note-taking, voice-transcription, obsidian-workflows] > author: [MISSING] > source: YouTube > reference: https://www.youtube.com/watch?v=2Jl6H9Ri8iI > date created: {{date:YYYY-MM-DD}} {{time:HH:mm}} > > # 🏷️ Title: My AI Note-taking workflows >  ## 🧭 Table of Contents - [📝 Summary](#-summary) - [📖 Main Content](#-main-content) - [🗒️ Notes](#️-notes) - [📚 References](#-references) - [🗣️ Paraphrase](#️-paraphrase) - [📰 Newsletter](#-newsletter) - [✅ Action Items](#-action-items) - [🧪 Hypotheses](#-hypotheses) - [📊 Methodology](#-methodology) - [❓ Questions](#-questions) - [🔮 Future Research](#-future-research) - [💡 Implications](#-implications) - [🧠 Learning Styles Assessment](#-learning-styles-assessment) - [📊 Accessibility Assessment](#-accessibility-assessment) - [🌍 DEIB Assessment](#-deib-assessment) - [✅ Logical Validity](#-logical-validity) - [🛡️ DARVO Analysis](#-darvo-analysis) - [🎯 Conclusion](#-conclusion) ## 📝 Summary - A note-taker describes their evolution from AI skeptic to power user, implementing three tiers of AI-powered voice-to-text workflows - Tier 1 (free): Local transcription apps like Spoken with offline models - Tier 2 (paid): Cloud services like Granola and Voice Notes with AI summaries and Q&A - Tier 3 (overkill): Hardware devices like Plaud for continuous life logging - The goal is reducing friction in note capture to focus on synthesis rather than transcription - Ethical concerns around recording others remain unresolved ## 📖 Main Content The speaker distinguishes between "note taking" (passive collection) and "note making" (synthesis and connection), arguing AI fits better in the former role. Their three-tier approach progresses from simple transcription to comprehensive life logging. **Tier 1: Free Local Transcription** Uses apps like Spoken with local models (Nvidia Parakeet) that work offline. The workflow involves holding a command key while speaking into any app, with instant transcription when released. Benefits include privacy and speed; limitations include lack of AI analysis. **Tier 2: Paid Cloud Services** Granola handles meetings with calendar integration, real-time Q&A, and automatic summaries. Voice Notes serves personal diary entries with cross-note queries. Both integrate with Obsidian through custom scripts and community plugins. Trade-offs include subscription costs and privacy concerns. **Tier 3: Hardware Life Logging** The Plaud Note Pin records up to 23 hours continuously. Transcripts feed into an OpenClaw AI agent with persistent memory for daily summaries. The speaker experiments with full-day recording during travel, creating automatic activity logs. ## 🗒️ Notes ### ⏱️ [00:00:00](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=0) Introduction and philosophy shift - Complete reversal from AI skepticism to heavy reliance - Focus on speech-to-text as primary use case ### ⏱️ [00:01:30](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=90) Nick Milo's distinction - Note taking = passive collection - Note making = synthesis and connection - AI better suited for note taking phase ### ⏱️ [00:02:15](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=135) Tier 1: Free solutions - iOS Voice Memos, Samsung Voice Recorder - Spoken app with local Nvidia Parakeet model - Command key trigger for instant transcription ### ⏱️ [00:05:30](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=330) Tier 2: Paid services - Granola for meetings with calendar integration - Voice Notes for personal journaling - Both offer AI summaries and Q&A capabilities ### ⏱️ [00:08:45](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=525) Obsidian integration - Custom macros and community plugins - Granola Sync and Voice Notes Sync plugins - Automated linking to daily notes ### ⏱️ [00:11:20](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=680) Tier 3: Plaud hardware - Wrist-worn device for continuous recording - Custom templates for different contexts - 23+ hour battery life for full-day logging ### ⏱️ [00:14:00](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=840) OpenClaw integration - AI agent with persistent memory - Processes daily transcripts into summaries - Captures forgotten details like restaurant names ### ⏱️ [00:16:30](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=990) Ethical considerations - Consent challenges in group conversations - Gray areas in conference settings - Ongoing personal navigation of boundaries ### ⏱️ [00:18:00](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=1080) Combined workflow benefits - Uses all three tiers for different contexts - Near-elimination of capture friction - Focus shifts to synthesis over transcription ## 📚 References - [Spoken app](https://spoken.ai/) - [Granola meeting notes](https://granola.so/) - [Voice Notes app](https://voicenotes.com/) - [Plaud Note Pin](https://plaud.ai/) - [Obsidian Granola Sync plugin](https://github.com/granola-so/obsidian-plugin) - [Nicole Vander Hooven's Plaud sync version](https://github.com/nicolevanderhoeven/plaud-sync-obsidian) ## 🗣️ Paraphrase A productivity enthusiast shares their comprehensive approach to AI-powered note-taking, moving from complete AI resistance to adopting multiple voice-to-text solutions. They've created a three-tier system: free local transcription for quick capture, paid cloud services for meetings and journaling, and hardware devices for continuous life logging. The goal is removing friction from note capture to focus mental energy on synthesis and connection-making rather than transcription drudgery. ## 📰 Newsletter Voice-powered note-taking is having a moment, and this deep-dive into three implementation tiers shows just how far the technology has come. From free local models that work offline to hardware devices that log your entire day, there's now a solution for every privacy comfort level and budget. The key insight? AI excels at the tedious capture phase, freeing you to focus on the uniquely human work of synthesis and meaning-making. If you've been hesitant about AI in your workflow, speech-to-text might be the perfect low-stakes entry point. ## ✅ Action Items #### DevRel - Developer advocacy: Demo voice-to-text integrations at WordPress meetups to show accessibility benefits - Education: Create workshop curriculum on implementing speech-to-text for content creators - Developer experience: Evaluate voice input plugins for wp-admin and Gutenberg editor - Developer marketing: Case study campaign featuring accessibility improvements through voice interfaces #### Community - Contributors: Voice-to-text onboarding videos for contributors with different accessibility needs - Maintainers: Explore voice note templates for code review feedback and issue triage #### WordPress - Accessibility: Audit current voice input support across Core, evaluate Web Speech API integration - Documentation: Voice-friendly content guidelines for HelpHub and DevHub - Gutenberg: Research voice-to-block conversion for faster content creation ## 🧪 Hypotheses - Local voice transcription reduces privacy barriers to AI adoption in note-taking workflows - Multi-tier implementation allows gradual adoption based on comfort and needs - Voice capture eliminates the friction that prevents many potential insights from being recorded - Hardware life logging creates more comprehensive personal knowledge bases than manual methods - AI excels at note-taking (capture) but humans remain superior at note-making (synthesis) ### 📊 Methodology - Comparative analysis across three implementation tiers (free/local, paid/cloud, hardware/continuous) - User experience evaluation through workflow demonstration - Integration testing with Obsidian knowledge management system - Cost-benefit analysis including privacy trade-offs - Longitudinal self-tracking of capture frequency and quality ### ❓ Questions - How do consent models scale for continuous recording in professional environments? - What's the optimal balance between capture comprehensiveness and processing overhead? - How does voice-first capture affect the quality of synthesized insights? - What are the long-term retention and searchability benefits of comprehensive voice logging? - How might voice-to-text workflows change collaborative knowledge work? ### 🔮 Future Research - Longitudinal studies on knowledge retention with voice-captured vs manually typed notes - Cross-cultural analysis of voice transcription accuracy across languages and accents - Privacy-preserving techniques for group conversation transcription - Integration patterns between voice capture and spaced repetition systems - Comparative analysis of synthesis quality from voice-captured vs traditional notes ### 💡 Implications - **Practice**: Voice-first workflows may become standard for knowledge workers - **Policy**: Need for clear consent frameworks around ambient recording in workplaces - **Security**: Local-first models reduce data exposure while maintaining functionality - **Education**: Speech-to-text democratizes note-taking for learners with different abilities - **Funding**: Hardware transcription devices represent new category of productivity tools ### 🧠 Learning Styles Assessment Reading level: 12th grade (Flesch-Kincaid). Technical vocabulary balanced with conversational explanations. #### 👀 Visual learners - Present: Screenshots of app interfaces, workflow diagrams, hardware photos - Improve: Side-by-side feature comparison tables, decision tree for tier selection #### 🔊 Aural learners - Present: Video format with clear narration, audio examples of transcription - Improve: Podcast version highlighting key workflow decisions #### 🗣️ Verbal learners - Present: Clear definitions of "note-taking vs note-making," step-by-step explanations - Improve: Glossary of technical terms, comparison analogies for different approaches #### 🧩 Kinesthetic learners - Present: Hands-on demonstrations, actual app usage examples - Improve: Setup checklists, trial period recommendations for each tier #### 🧠 Logical learners - Present: Three-tier progression, cost-benefit analysis, integration workflows - Improve: Decision matrix for tier selection, ROI calculations for paid solutions #### 🧑‍🤝‍🧑 Social learners - Present: Community plugin mentions, collaboration considerations - Improve: Discussion prompts about ethical boundaries, group adoption strategies #### 🧘‍♀️ Solitary learners - Present: Personal workflow focus, individual productivity gains - Improve: Self-assessment questions for determining optimal tier, reflection prompts #### 🌿 Naturalistic learners - Present: Organic workflow evolution, natural speech patterns - Improve: Environmental impact considerations of cloud vs local processing ### 📊 Accessibility Assessment #### Screen reader semantics - Present: Clear heading hierarchy, descriptive video narration - Improve: Alt text for interface screenshots, transcript with semantic markup #### Keyboard navigation - Present: Keyboard shortcuts for voice activation mentioned - Improve: Full keyboard alternatives for all demonstrated workflows #### Contrast and readability - Present: Clear video quality, readable interface examples - Improve: High-contrast mode compatibility for recommended apps #### Text resizing and reflow - Present: Video format accommodates different viewing sizes - Improve: Responsive transcript layout, scalable UI recommendations #### Media and captions - Present: Spoken narration with clear audio - Improve: Closed captions, downloadable transcript with timestamps #### Forms and interactions - Present: Voice activation reduces traditional form interaction needs - Improve: Fallback methods for users who cannot use voice input ### 🌍 DEIB Assessment #### Cultural breadth - Present: Acknowledgment of different device ecosystems (iOS/Android) - Improve: Discussion of accent and language support across transcription models #### Inclusive language - Present: Gender-neutral references, avoids ableist assumptions - Improve: Recognition that voice input benefits extend beyond convenience #### Neurodiversity support - Present: Multiple capture methods accommodate different processing preferences - Improve: Explicit discussion of ADHD, executive function benefits #### Representation - Present: Single perspective limits representation - Improve: Multiple user stories showing diverse implementation approaches #### Psychological safety - Present: Ethical considerations section shows awareness of consent issues - Improve: Clearer guidelines for respectful recording practices ### ✅ Logical Validity #### Strong reasoning - **Tiered approach**: Logically progresses from simple to complex, allowing graduated adoption - **Use case matching**: Different tools mapped to specific contexts (meetings, journaling, life logging) - **Integration focus**: Demonstrates how AI enhances rather than replaces human synthesis #### Fallacies - **Appeal to novelty**: Implies newer AI tools are inherently better without comparing to refined manual methods - **False dichotomy**: Presents choice as either manual note-taking or AI assistance, ignoring hybrid approaches - **Anecdotal evidence**: Single user's experience may not generalize to different workflows or needs #### Tone of voice Professional enthusiasm with personal anecdotes. Uses accessible language while maintaining technical accuracy. ### 🛡️ DARVO Analysis No DARVO patterns detected. The content acknowledges ethical concerns and personal limitations rather than deflecting criticism. ## 🎯 Conclusion This comprehensive workflow demonstrates how AI can dramatically reduce friction in knowledge capture while preserving human agency in synthesis and meaning-making. The three-tier approach provides clear entry points for different comfort levels with AI and privacy trade-offs. Most importantly, the focus remains on enhancing human cognition rather than replacing it—using AI to eliminate transcription drudgery so mental energy can flow toward uniquely human activities like connection-making and insight generation. ## 🔄 Refactor Appendix — My AI Note-taking workflows ### 📚 Literature Atoms (H3) #### Voice transcription eliminates capture friction better than manual typing > [!quote] ID:rw_manual_transcription > The friction between thinking of something and writing it down is basically gone for me and I've always been a note taker, but I still didn't take notes on everything. Now it's still not everything, but it's pretty damn close. **Context & use**: - Core insight from implementing multi-tier voice workflow - Quantifies impact on capture completeness vs manual methods **Tags**: #xref #productivity-workflows **Highlight**: Manual transcription remains a significant barrier to comprehensive knowledge capture  #### AI suits note-taking better than note-making for synthesis preservation > [!quote] ID:rw_milo_distinction > Nick Milo makes a distinction between note taking, which is the passive more collecting aspect where you're just trying to write down what you heard or read and note making, which is connecting notes, linking them to each other and synthesizing and deriving insights from them. I think that AI naturally fits better for me at least in the note-taking aspect. **Context & use**: - Philosophical foundation for AI integration strategy - Preserves human agency in synthesis while automating capture **Tags**: #xref #ai-human-collaboration **Highlight**: Clear boundary between AI automation and human creativity  #### Local models reduce privacy barriers to AI adoption > [!quote] ID:rw_local_privacy > What I like about this is it's really quick. There's no data that's being sent on the internet, so this is gonna work. Even if you have no internet connection. **Context & use**: - Tier 1 implementation using Spoken app with local models - Privacy-first approach enables broader AI acceptance **Tags**: #xref #privacy-preserving-ai **Highlight**: Local processing maintains control while enabling AI benefits  ### 🧠 Permanent Candidates (H4) #### Graduated AI adoption reduces implementation friction through tiered complexity **Claim**: Three-tier systems (free/local, paid/cloud, hardware/continuous) enable progressive AI adoption by matching complexity to user readiness and specific use cases. **Evidence pointer**: [[#Voice transcription eliminates capture friction better than manual typing]] **Scope & conditions**: - Holds when users have varying privacy comfort levels and technical skills - Breaks when forced migration between tiers due to service changes **Links**: - [[Progressive disclosure in UX]] — because applies-in software adoption patterns - [[Privacy-utility trade-offs]] — because extends decision framework - [[Technology adoption curves]] — because evidence-for graduated implementation  ### ✏️ Fleeting Queue - F1: Research consent models for ambient recording in professional settings — ethics implications for workplace AI - F2: Evaluate WordPress voice input plugins for accessibility improvements — Core integration potential ### → Things 3 - [ ] Test local transcription accuracy across multiple languages for international community support #things - [ ] Draft ethical guidelines template for voice recording in collaborative environments #things --- ## Full Transcript [00:00] - Just over a year ago, if you had asked me [00:01] how I felt about AI [00:03] and note taking, I would've said that there's no way [00:06] that I wanted AI near my notes. [00:09] Fast forward a year later [00:10] and the world is different and so am I. [00:13] So here I am at a point where I heavily rely on AI [00:17] to take my notes in many different ways [00:20] of the many ways that I use ai. [00:22] Now, I think probably the easiest to start with is speech [00:25] to text, meaning taking my voice [00:28] and turning that into plain text [00:31] that I can store in my obsidian vault. [00:33] In this video, I'm gonna show you three different levels [00:36] that I've implemented this at and my workflows for each one. [00:41] Here's the way that I'm currently thinking [00:42] of AI's place in my notes Nick Milo makes a distinction [00:46] between note taking, [00:47] which is the passive more collecting aspect [00:50] where you're just trying to write down what you heard [00:52] or read and note making, which is connecting notes, [00:57] linking them to each other and synthesizing [01:00] and deriving insights from them. [01:02] I think that AI naturally fits better [01:05] for me at least in the note-taking aspect. [01:09] That way I am kind of outsourcing a lot of the tedium [01:12] of taking down notes in the first place [01:14] and not so much the synthesizing [01:17] and creating new meaning from my notes. [01:21] The first tier is completely free, [01:23] so no credit card information required for either the apps [01:26] or the underlying LLMs that those apps are using. [01:31] I think the best way to start is to just use your phone. [01:34] So both iOS and Android have good alternatives for this. [01:38] iOS already transcribes voice recordings on the Voice memos [01:42] app and there is an alternative for Samsung devices as well. [01:47] At least if you have one of the later Galaxy devices, [01:50] then you're gonna be able [01:51] to use the voice recorder app to do the same thing. [01:55] The one that I actually use the most though is [01:57] something called Spoken. [01:59] It is a free app and I'm looking at it right now. [02:02] When you start the app, [02:03] you can look at the dictation models here. [02:07] If you see down here I, I'm using a free account, [02:10] you can upgrade to pro if you want to use some [02:12] of the frontier models, but I'm not using it for that. [02:16] Instead I'm using this local model, [02:20] it's called Nvidia Parakeet [02:22] and so you can choose between a lot of different ones. [02:25] Apparently there's a newer version of the one [02:28] that I downloaded that is supposedly faster. [02:31] There's also some Quinn ones, [02:33] so these are free what's called open weight local models [02:37] that you can download and you, [02:39] and they're the ones that do the actual transcription. [02:42] So how this works is you go into any app, [02:45] let's say obsidian, [02:46] but you know it doesn't actually have to be obsidian, [02:48] it could be like in your browser. [02:49] So you could dictate an email. [02:52] For instance, I'm gonna use obsidian here [02:56] and I'm just going to create a new note and all I have to do [03:00] and you can change what key you press, [03:03] I'm pressing down the command key [03:05] and then I can just start typing. [03:07] You can see that there's something at the bottom there from [03:10] spoken that notices that I'm holding this down [03:13] and that I'm speaking into obsidian [03:16] and then I'm gonna let go [03:18] of the right command key and there it is. [03:20] Everything I've just said is now text in obsidian. [03:24] What I like about this is it's really quick. [03:26] There's no data that's being sent on the internet, [03:29] so this is gonna work. [03:30] Even if you have no internet connection [03:33] and you just hold down a command key [03:36] that it's pretty easy to trigger it. [03:38] You can also change it so that it's more of a toggle thing, [03:41] so you press once to start the recording [03:44] and then you press again to stop it. [03:46] But I find myself using Spoken Lee more for like quick sort [03:50] of thing, so I just like to hold it down anyway. [03:54] Now, this free tier is what I would recommend if you kind [03:57] of just want to try it out [03:59] and see if this will make it into your workflows. [04:02] This is I think, the most accessible one [04:05] because you don't have to pay any money. [04:08] What it's not so good for is longer content. [04:11] You know, you can do the toggle thing like I mentioned, [04:14] but there is no asking AI about [04:18] what you've just recorded. [04:19] There's really just translating from words that you, [04:24] you speak to, words that you've written and that's it. [04:27] Sometimes that's enough, [04:29] but if it's not, then we go into tier two, [04:31] which is the paid tier. [04:33] In tier two, you're basically paying [04:35] for the convenience layer. [04:37] So in tier two you have these paid apps which often have a [04:41] free component, [04:42] but I've found that it's really more realistic to think [04:46] of it as paid because the free tiers are not that generous. [04:50] These are tools that you might already be familiar [04:52] with if you're having meetings online. [04:55] So Google Meet for example has Gemini that can record [04:59] and transcribe what is being said. [05:01] Zoom has their AI companion as well. [05:04] Auto AI is one that's been out for a while, [05:07] but the ones that I use, there are two [05:09] of them are voice notes and granola. [05:12] Now granola is really good for meetings [05:16] and I really like that here in granola. [05:19] It brings in my calendar, so I've linked it [05:22] to my Google calendar so it's aware of the meetings [05:24] that come up and [05:25] whenever I start one of them, [05:27] a little notification comes up in the corner that says, [05:30] do you wanna take notes? [05:31] And it's very easy to just say yes, take notes. [05:34] This is actually the summary [05:36] that was generated from a video I watched [05:38] by a YouTuber called Book Born. [05:40] My partner and I are wanting [05:42] to read the latest engine crawler Carl book, [05:46] and we've already forgotten the previous seven books. [05:48] So I watched her recap of it [05:50] and this is the notes that granola automatically took of it. [05:56] A really cool thing here is that [05:57] with granola you can ask questions live. [06:01] So I obviously did this already and it's not live, [06:05] but I can still kind of show you what that looks like. [06:08] So I'll say, who is Princess Donut? [06:11] And I'll enter that and in real time or apparently [06:15] after the fact it can answer questions. [06:17] So now it's saying Princess Donut is Carl Sapien Cat [06:20] companion and one [06:21] of the two central protagonists in the book. [06:23] In meetings, I actually find this really useful if I get [06:26] distracted, especially in long meetings [06:29] and I'm like, can you just recap the last five minutes [06:33] because I was not paying attention. [06:35] Let me flip back to obsidian. [06:37] This is my actual obsidian note for it. [06:39] This is the video that I was watching. [06:42] I created a macro, so now I just have [06:45] to open up the command pane and then, and I type granola [06:50] and I have this link granola one, I hit enter [06:53] and then I can choose which one it was. [06:55] So in this case it was Dun crawler Carl. [06:59] So now the granola section is added [07:02] to the bottom of the note. [07:03] So I, I set all of this up in the macro [07:06] and now that summary is here as well. [07:10] So every time I open the obsidian note, [07:13] the entire summary is also there. [07:15] And then at the bottom here, you can also look at exactly [07:19] what was being said there [07:21] and it's like nicely formatted too. [07:25] So this is using two things. [07:28] One is the granola plugin, it's called, yeah, granola sync. [07:33] It is a community plugin. [07:35] And then it's also using a specific script that I wrote just [07:39] to be able to link granola transcripts [07:42] and summaries to the notes, [07:44] to the corresponding notes in obsidian. [07:46] And you can ask your AI to write yourself a script [07:49] that works for you. [07:50] If you want a copy of mine in particular, [07:53] I do have it in my Patreon vault, [07:55] so it is also here, granola. [07:58] The other thing that I use is voice notes. [08:00] Now I tend to use voice notes actually mostly on my phone. [08:04] My use case of voice notes is actually a little too personal [08:07] to show because it's more for like diary style entry. [08:11] So I can't show you the interface, [08:13] but you can also ask the AI there for certain things [08:18] that span different notes. [08:20] So how many times have I talked about this or, [08:23] or when was the last time that I talked about this person? [08:26] So that kind of thing. That requires multiple notes. [08:28] It's pretty good at retrieving [08:30] and on obsidian there's also a voice note sync app [08:34] that works really well. [08:35] And just like you saw with granola, I do also have a macro [08:40] that runs a script for linking voice notes to my daily notes [08:44] because my use case of it is more like a journal. [08:47] So I want it always embedded in my daily notes. [08:50] Some trade-offs on this tier is, well the cost [08:54] and also the fact [08:55] that you are losing a little bit in privacy [08:57] because you are now sending those notes to other people [09:00] to store on their servers. [09:02] There's also the risk of [09:04] that whole workflow getting interrupted if that [09:07] that subscription service is no longer functional. [09:10] So you do take a little bit more risk with it. [09:13] What you gain though is a lot more convenience. [09:17] I find that I do voice notes [09:18] and granola actually more than I use openly, [09:22] and I think it's just because it's, it's smarter. [09:25] So I think that if you're someone [09:27] who has something specific in mind to capture audio [09:29] for like a talk at a conference or a video or a voice memo [09:34] and you're okay with starting [09:36] and stopping it, then I think the tools in this [09:39] section are for you. [09:40] If you want even more, then there's like tier three, [09:45] which is the overkill, completely ridiculous level [09:50] that I'm gonna show you a workflow for anyway, [09:52] because I am that ridiculous. [09:55] Let me introduce you to plot. [09:57] You might not have noticed this on my wrist, [10:00] but it is a little hardware recorder. [10:03] It is something that sits on your wrist here. [10:06] It doesn't come with this band. [10:08] This is something that I got off market, [10:10] but it comes with a much uglier band [10:13] or it comes with a necklace or a pin [10:16] and I kind of mix it up a little based on [10:20] how obtrusive or unobtrusive I want it to be. [10:24] It is basically a recorder [10:27] and you can just use it as a recorder. [10:30] So you can just have it as something that you wear [10:33] that records your voice, [10:34] and then at the end of the day you can transfer all [10:37] of those recordings to your laptop [10:39] or your mobile device where you can then use AI [10:43] to transcribe it [10:45] or you can optionally send those recordings to their cloud. [10:50] So what I do here is I use this plot sync plugin, [10:53] which I did not create. [10:55] It is by Leonard sell, [10:57] but I found that there were some things that I wanted [11:00] to change that I wasn't sure [11:02] that everybody would really want. [11:04] So you can download his version and install it. [11:08] Or if you want, you can also download my version of it, [11:12] which is on Nicole Vander Holosync for obsidian. [11:16] I'll put a link in the description [11:18] and it just formats it a little better [11:20] for this particular use case. [11:23] It works for me. [11:24] You know, you, I would encourage you to choose the version [11:27] that works best for you. [11:28] So that is the obsidian part of it. [11:30] That's how I get these transcripts into obsidian [11:34] and when they come through, [11:35] they're already nicely formatted and summarized. [11:38] One cool thing that I like about plot that I don't get [11:41] with granola or voice notes is [11:43] that plot lets you have templates [11:46] so they're not always just processed in a certain way. [11:50] Sometimes I've recorded like games, for example, [11:53] and when I'm playing a role playing game, [11:55] I do not wanna have like next actions [11:58] and agenda items the way I would in a meeting. [12:01] So I have a game one and I have a meeting template [12:05] and then I choose the more relevant template for [12:07] what I'm recording. [12:09] But one thing that I've actually been experimenting [12:11] with is just using it for life logging. [12:15] So turning it on at the beginning of the day [12:18] and letting it just record everything because PLA devices, [12:22] or at least this PLA note pin, there are a few kinds [12:25] of devices including like a card shaped one, [12:29] but they can record for like up to 23 hours or more I think. [12:34] So I've also been experimenting with using it [12:36] as a life logging tool. [12:38] I really love to travel and my partner [12:40] and I are really enjoying seeing new parts of Portugal. [12:45] And so with his consent, [12:47] obviously I have recorded the entire day [12:50] of our conversations [12:52] and those conversations I then feed to my open claw agent. [12:57] Now, I'm not gonna get into open claw here, [12:59] but basically it is, think of it as an AI assistant [13:02] with a persistent memory. [13:04] So this is something that knows you [13:06] and has an extent, a much more extensive memory of [13:11] what you like and the things that you're interested in. [13:14] And it's also a gentech, [13:15] so it can do things like create notes. [13:18] So I'm gonna show you just briefly how [13:21] that translates in my obsidian vault. [13:24] This is my daily note for May 2nd, [13:27] and you can see that I have a section here for plot. [13:31] That's because I do have macro for that as well. [13:35] And it is this one, this linking plot to the daily note. [13:39] Each individual one looks like this. [13:41] So that this one is I, I was using a template [13:44] where it's really just a transcript and no summaries [13:48] and it has these timestamps here for the raw transcript. [13:51] Now my open claw agent Iris, [13:54] and if you wanna know why I call her Iris, [13:56] then check out this video up at the top [14:00] where I talk a little bit more about that setup. [14:02] Looks through all of the plot transcripts. [14:04] So in this case it was 15 hours of recording time or or less [14:09] 'cause it's five hours per recording. [14:12] And she created my open class assistant, [14:15] created a daily summary of it. [14:17] So talks about where we were, the RA coast [14:20] and a restaurant that we went to. [14:23] She basically took my voice transcripts [14:27] and turned it into a summary of the day. [14:30] These are things that I now don't remember, [14:32] like do I remember that it was called the mesino? [14:35] No. All right. This is not without its cons. [14:40] First, it has to be addressed. [14:41] There are real ethical concerns here. [14:44] I try not to record people, especially if they say [14:48] that they don't wanna be recorded. [14:49] That's no questions. I will stop that for you. [14:53] But the problem is that there are all sorts of gray areas [14:57] where, you know what if I'm in a conference [14:59] and I'm speaking to someone [15:00] and that person agrees to be recorded, [15:03] but another person just suddenly comes up [15:06] and starts talking, that third person is now being recorded. [15:10] So do I then have an obligation to stop [15:14] and say like, Hey, we're recording this. [15:15] Is that still okay? [15:17] That can be unwieldy in a conference setting, for example, [15:21] which is actually what I got it for so that I can remember [15:24] who said what without actually stopping and taking notes. [15:29] So I'm still trying to figure out what, where I am on [15:33] that spectrum and you can figure that out [15:36] for yourself if, if you want that too. [15:38] One of the benefits is that I no longer have [15:41] to stop at the end of the day and think, where did I go? [15:44] What was the name of that restaurant? [15:46] Now it just is kind of passively recorded. [15:50] At the end of the day, I still have a not perfect [15:54] but way better record of what I've done than [15:57] what I normally have time for. [16:00] So regardless of which of the three levels you choose [16:03] or if you choose any at all, what I am getting out [16:06] of this is that I think voice is now the least friction [16:10] capture method that exists for me [16:13] because I get to just do the thing and live my life [16:16] and have it passively recorded [16:18] and my tasks waiting for me when I get home back [16:21] to my obsidian vault. [16:22] The real overkill thing here is [16:24] that I still use all three levels. [16:26] I use spoken for just writing an email [16:30] or dictating something into obsidian. [16:32] I use voice notes for more personal diary type entries, [16:36] granola for meetings, videos and games. [16:40] And I use plot as my kind of daily life lager [16:44] and capture of things that I didn't think to capture. [16:48] The result of implementing these tools is that for me, [16:52] the friction between thinking of something [16:54] and writing it down is basically gone for me [16:57] and I've always been a note taker, [16:59] but I still didn't take notes on everything. [17:02] Now it's still not everything, but it's pretty damn close. [17:06] And that means that if I can outsource that part of it, [17:10] then I get to focus on being present, staying in the moment, [17:14] and if I choose to take notes, it's not going to be [17:17] what I heard or what happened [17:19] or anything surface level like that. [17:21] I can focus on the actual synthesis. [17:25] So I'm getting to that connecting, [17:27] synthesizing part earlier on in the cycle. [17:31] If you want any more details on any of these workflows, [17:34] then drop something in the comment below so that I know [17:37] that I should make a video on that.  ## 📺 Video ![](https://www.youtube.com/watch?v=2Jl6H9Ri8iI) ## Highlights - The friction between thinking of something and writing it down is basically gone for me and I've always been a note taker, but I still didn't take notes on everything. Now it's still not everything, but it's pretty damn close. - Note: 📚 Voice transcription eliminates capture friction better than manual typing · - Core insight from implementing multi-tier voice workflow - Quantifies impact on capture completeness vs manual methods · #atom/literature/h1 - Nick Milo makes a distinction between note taking, which is the passive more collecting aspect where you're just trying to write down what you heard or read and note making, which is connecting notes, linking them to each other and synthesizing and deriving insights from them. I think that AI naturally fits better for me at least in the note-taking aspect. - Note: 📚 AI suits note-taking better than note-making for synthesis preservation · - Philosophical foundation for AI integration strategy - Preserves human agency in synthesis while automating capture · #atom/literature/h2 - What I like about this is it's really quick. There's no data that's being sent on the internet, so this is gonna work. Even if you have no internet connection. - Note: 📚 Local models reduce privacy barriers to AI adoption · - Tier 1 implementation using Spoken app with local models - Privacy-first approach enables broader AI acceptance · #atom/literature/h3 - Graduated AI adoption reduces implementation friction through tiered complexity - Note: Holds when users have varying privacy comfort levels and technical skills · - [[Progressive disclosure in UX]] — because applies-in software adoption patterns · #atom/zettel/z1 --- # 📝 Full Notes (Deep-Read Analysis) # 🏷️ Title: My AI Note-taking workflows ## 🧭 Table of Contents - [📝 Summary](#-summary) - [📖 Main Content](#-main-content) - [🗒️ Notes](#️-notes) - [📚 References](#-references) - [🗣️ Paraphrase](#️-paraphrase) - [📰 Newsletter](#-newsletter) - [✅ Action Items](#-action-items) - [🧪 Hypotheses](#-hypotheses) - [📊 Methodology](#-methodology) - [❓ Questions](#-questions) - [🔮 Future Research](#-future-research) - [💡 Implications](#-implications) - [🧠 Learning Styles Assessment](#-learning-styles-assessment) - [📊 Accessibility Assessment](#-accessibility-assessment) - [🌍 DEIB Assessment](#-deib-assessment) - [✅ Logical Validity](#-logical-validity) - [🛡️ DARVO Analysis](#-darvo-analysis) - [🎯 Conclusion](#-conclusion) ## 📝 Summary - A note-taker describes their evolution from AI skeptic to power user, implementing three tiers of AI-powered voice-to-text workflows - Tier 1 (free): Local transcription apps like Spoken with offline models - Tier 2 (paid): Cloud services like Granola and Voice Notes with AI summaries and Q&A - Tier 3 (overkill): Hardware devices like Plaud for continuous life logging - The goal is reducing friction in note capture to focus on synthesis rather than transcription - Ethical concerns around recording others remain unresolved ## 📖 Main Content The speaker distinguishes between "note taking" (passive collection) and "note making" (synthesis and connection), arguing AI fits better in the former role. Their three-tier approach progresses from simple transcription to comprehensive life logging. **Tier 1: Free Local Transcription** Uses apps like Spoken with local models (Nvidia Parakeet) that work offline. The workflow involves holding a command key while speaking into any app, with instant transcription when released. Benefits include privacy and speed; limitations include lack of AI analysis. **Tier 2: Paid Cloud Services** Granola handles meetings with calendar integration, real-time Q&A, and automatic summaries. Voice Notes serves personal diary entries with cross-note queries. Both integrate with Obsidian through custom scripts and community plugins. Trade-offs include subscription costs and privacy concerns. **Tier 3: Hardware Life Logging** The Plaud Note Pin records up to 23 hours continuously. Transcripts feed into an OpenClaw AI agent with persistent memory for daily summaries. The speaker experiments with full-day recording during travel, creating automatic activity logs. ## 🗒️ Notes ### ⏱️ [00:00:00](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=0) Introduction and philosophy shift - Complete reversal from AI skepticism to heavy reliance - Focus on speech-to-text as primary use case ### ⏱️ [00:01:30](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=90) Nick Milo's distinction - Note taking = passive collection - Note making = synthesis and connection - AI better suited for note taking phase ### ⏱️ [00:02:15](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=135) Tier 1: Free solutions - iOS Voice Memos, Samsung Voice Recorder - Spoken app with local Nvidia Parakeet model - Command key trigger for instant transcription ### ⏱️ [00:05:30](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=330) Tier 2: Paid services - Granola for meetings with calendar integration - Voice Notes for personal journaling - Both offer AI summaries and Q&A capabilities ### ⏱️ [00:08:45](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=525) Obsidian integration - Custom macros and community plugins - Granola Sync and Voice Notes Sync plugins - Automated linking to daily notes ### ⏱️ [00:11:20](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=680) Tier 3: Plaud hardware - Wrist-worn device for continuous recording - Custom templates for different contexts - 23+ hour battery life for full-day logging ### ⏱️ [00:14:00](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=840) OpenClaw integration - AI agent with persistent memory - Processes daily transcripts into summaries - Captures forgotten details like restaurant names ### ⏱️ [00:16:30](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=990) Ethical considerations - Consent challenges in group conversations - Gray areas in conference settings - Ongoing personal navigation of boundaries ### ⏱️ [00:18:00](https://www.youtube.com/watch?v=2Jl6H9Ri8iI&t=1080) Combined workflow benefits - Uses all three tiers for different contexts - Near-elimination of capture friction - Focus shifts to synthesis over transcription ## 📚 References - [Spoken app](https://spoken.ai/) - [Granola meeting notes](https://granola.so/) - [Voice Notes app](https://voicenotes.com/) - [Plaud Note Pin](https://plaud.ai/) - [Obsidian Granola Sync plugin](https://github.com/granola-so/obsidian-plugin) - [Nicole Vander Hooven's Plaud sync version](https://github.com/nicolevanderhoeven/plaud-sync-obsidian) ## 🗣️ Paraphrase A productivity enthusiast shares their comprehensive approach to AI-powered note-taking, moving from complete AI resistance to adopting multiple voice-to-text solutions. They've created a three-tier system: free local transcription for quick capture, paid cloud services for meetings and journaling, and hardware devices for continuous life logging. The goal is removing friction from note capture to focus mental energy on synthesis and connection-making rather than transcription drudgery. ## 📰 Newsletter Voice-powered note-taking is having a moment, and this deep-dive into three implementation tiers shows just how far the technology has come. From free local models that work offline to hardware devices that log your entire day, there's now a solution for every privacy comfort level and budget. The key insight? AI excels at the tedious capture phase, freeing you to focus on the uniquely human work of synthesis and meaning-making. If you've been hesitant about AI in your workflow, speech-to-text might be the perfect low-stakes entry point. ## ✅ Action Items #### DevRel - Developer advocacy: Demo voice-to-text integrations at WordPress meetups to show accessibility benefits - Education: Create workshop curriculum on implementing speech-to-text for content creators - Developer experience: Evaluate voice input plugins for wp-admin and Gutenberg editor - Developer marketing: Case study campaign featuring accessibility improvements through voice interfaces #### Community - Contributors: Voice-to-text onboarding videos for contributors with different accessibility needs - Maintainers: Explore voice note templates for code review feedback and issue triage #### WordPress - Accessibility: Audit current voice input support across Core, evaluate Web Speech API integration - Documentation: Voice-friendly content guidelines for HelpHub and DevHub - Gutenberg: Research voice-to-block conversion for faster content creation ## 🧪 Hypotheses - Local voice transcription reduces privacy barriers to AI adoption in note-taking workflows - Multi-tier implementation allows gradual adoption based on comfort and needs - Voice capture eliminates the friction that prevents many potential insights from being recorded - Hardware life logging creates more comprehensive personal knowledge bases than manual methods - AI excels at note-taking (capture) but humans remain superior at note-making (synthesis) ### 📊 Methodology - Comparative analysis across three implementation tiers (free/local, paid/cloud, hardware/continuous) - User experience evaluation through workflow demonstration - Integration testing with Obsidian knowledge management system - Cost-benefit analysis including privacy trade-offs - Longitudinal self-tracking of capture frequency and quality ### ❓ Questions - How do consent models scale for continuous recording in professional environments? - What's the optimal balance between capture comprehensiveness and processing overhead? - How does voice-first capture affect the quality of synthesized insights? - What are the long-term retention and searchability benefits of comprehensive voice logging? - How might voice-to-text workflows change collaborative knowledge work? ### 🔮 Future Research - Longitudinal studies on knowledge retention with voice-captured vs manually typed notes - Cross-cultural analysis of voice transcription accuracy across languages and accents - Privacy-preserving techniques for group conversation transcription - Integration patterns between voice capture and spaced repetition systems - Comparative analysis of synthesis quality from voice-captured vs traditional notes ### 💡 Implications - **Practice**: Voice-first workflows may become standard for knowledge workers - **Policy**: Need for clear consent frameworks around ambient recording in workplaces - **Security**: Local-first models reduce data exposure while maintaining functionality - **Education**: Speech-to-text democratizes note-taking for learners with different abilities - **Funding**: Hardware transcription devices represent new category of productivity tools ### 🧠 Learning Styles Assessment Reading level: 12th grade (Flesch-Kincaid). Technical vocabulary balanced with conversational explanations. #### 👀 Visual learners - Present: Screenshots of app interfaces, workflow diagrams, hardware photos - Improve: Side-by-side feature comparison tables, decision tree for tier selection #### 🔊 Aural learners - Present: Video format with clear narration, audio examples of transcription - Improve: Podcast version highlighting key workflow decisions #### 🗣️ Verbal learners - Present: Clear definitions of "note-taking vs note-making," step-by-step explanations - Improve: Glossary of technical terms, comparison analogies for different approaches #### 🧩 Kinesthetic learners - Present: Hands-on demonstrations, actual app usage examples - Improve: Setup checklists, trial period recommendations for each tier #### 🧠 Logical learners - Present: Three-tier progression, cost-benefit analysis, integration workflows - Improve: Decision matrix for tier selection, ROI calculations for paid solutions #### 🧑‍🤝‍🧑 Social learners - Present: Community plugin mentions, collaboration considerations - Improve: Discussion prompts about ethical boundaries, group adoption strategies #### 🧘‍♀️ Solitary learners - Present: Personal workflow focus, individual productivity gains - Improve: Self-assessment questions for determining optimal tier, reflection prompts #### 🌿 Naturalistic learners - Present: Organic workflow evolution, natural speech patterns - Improve: Environmental impact considerations of cloud vs local processing ### 📊 Accessibility Assessment #### Screen reader semantics - Present: Clear heading hierarchy, descriptive video narration - Improve: Alt text for interface screenshots, transcript with semantic markup #### Keyboard navigation - Present: Keyboard shortcuts for voice activation mentioned - Improve: Full keyboard alternatives for all demonstrated workflows #### Contrast and readability - Present: Clear video quality, readable interface examples - Improve: High-contrast mode compatibility for recommended apps #### Text resizing and reflow - Present: Video format accommodates different viewing sizes - Improve: Responsive transcript layout, scalable UI recommendations #### Media and captions - Present: Spoken narration with clear audio - Improve: Closed captions, downloadable transcript with timestamps #### Forms and interactions - Present: Voice activation reduces traditional form interaction needs - Improve: Fallback methods for users who cannot use voice input ### 🌍 DEIB Assessment #### Cultural breadth - Present: Acknowledgment of different device ecosystems (iOS/Android) - Improve: Discussion of accent and language support across transcription models #### Inclusive language - Present: Gender-neutral references, avoids ableist assumptions - Improve: Recognition that voice input benefits extend beyond convenience #### Neurodiversity support - Present: Multiple capture methods accommodate different processing preferences - Improve: Explicit discussion of ADHD, executive function benefits #### Representation - Present: Single perspective limits representation - Improve: Multiple user stories showing diverse implementation approaches #### Psychological safety - Present: Ethical considerations section shows awareness of consent issues - Improve: Clearer guidelines for respectful recording practices ### ✅ Logical Validity #### Strong reasoning - **Tiered approach**: Logically progresses from simple to complex, allowing graduated adoption - **Use case matching**: Different tools mapped to specific contexts (meetings, journaling, life logging) - **Integration focus**: Demonstrates how AI enhances rather than replaces human synthesis #### Fallacies - **Appeal to novelty**: Implies newer AI tools are inherently better without comparing to refined manual methods - **False dichotomy**: Presents choice as either manual note-taking or AI assistance, ignoring hybrid approaches - **Anecdotal evidence**: Single user's experience may not generalize to different workflows or needs #### Tone of voice Professional enthusiasm with personal anecdotes. Uses accessible language while maintaining technical accuracy. ### 🛡️ DARVO Analysis No DARVO patterns detected. The content acknowledges ethical concerns and personal limitations rather than deflecting criticism. ## 🎯 Conclusion This comprehensive workflow demonstrates how AI can dramatically reduce friction in knowledge capture while preserving human agency in synthesis and meaning-making. The three-tier approach provides clear entry points for different comfort levels with AI and privacy trade-offs. Most importantly, the focus remains on enhancing human cognition rather than replacing it—using AI to eliminate transcription drudgery so mental energy can flow toward uniquely human activities like connection-making and insight generation. ## 🔄 Refactor Appendix — My AI Note-taking workflows ### 📚 Literature Atoms (H3) #### Voice transcription eliminates capture friction better than manual typing > [!quote] ID:rw_manual_transcription > The friction between thinking of something and writing it down is basically gone for me and I've always been a note taker, but I still didn't take notes on everything. Now it's still not everything, but it's pretty damn close. **Context & use**: - Core insight from implementing multi-tier voice workflow - Quantifies impact on capture completeness vs manual methods **Tags**: #xref #productivity-workflows **Highlight**: Manual transcription remains a significant barrier to comprehensive knowledge capture  #### AI suits note-taking better than note-making for synthesis preservation > [!quote] ID:rw_milo_distinction > Nick Milo makes a distinction between note taking, which is the passive more collecting aspect where you're just trying to write down what you heard or read and note making, which is connecting notes, linking them to each other and synthesizing and deriving insights from them. I think that AI naturally fits better for me at least in the note-taking aspect. **Context & use**: - Philosophical foundation for AI integration strategy - Preserves human agency in synthesis while automating capture **Tags**: #xref #ai-human-collaboration **Highlight**: Clear boundary between AI automation and human creativity  #### Local models reduce privacy barriers to AI adoption > [!quote] ID:rw_local_privacy > What I like about this is it's really quick. There's no data that's being sent on the internet, so this is gonna work. Even if you have no internet connection. **Context & use**: - Tier 1 implementation using Spoken app with local models - Privacy-first approach enables broader AI acceptance **Tags**: #xref #privacy-preserving-ai **Highlight**: Local processing maintains control while enabling AI benefits  ### 🧠 Permanent Candidates (H4) #### Graduated AI adoption reduces implementation friction through tiered complexity **Claim**: Three-tier systems (free/local, paid/cloud, hardware/continuous) enable progressive AI adoption by matching complexity to user readiness and specific use cases. **Evidence pointer**: [[#Voice transcription eliminates capture friction better than manual typing]] **Scope & conditions**: - Holds when users have varying privacy comfort levels and technical skills - Breaks when forced migration between tiers due to service changes **Links**: - [[Progressive disclosure in UX]] — because applies-in software adoption patterns - [[Privacy-utility trade-offs]] — because extends decision framework - [[Technology adoption curves]] — because evidence-for graduated implementation  ### ✏️ Fleeting Queue - F1: Research consent models for ambient recording in professional settings — ethics implications for workplace AI - F2: Evaluate WordPress voice input plugins for accessibility improvements — Core integration potential ### → Things 3 - [ ] Test local transcription accuracy across multiple languages for international community support #things - [ ] Draft ethical guidelines template for voice recording in collaborative environments #things --- # 📄 Full Transcript *Auto-generated from YouTube captions · timestamps roughly every 30 seconds · [watch on YouTube](https://www.youtube.com/watch?v=2Jl6H9Ri8iI)* **[00:00]** Just over a year ago, if you had asked me how I felt about AI and note-taking, I would have said that there's no way [music] that I wanted AI near my notes. Fast forward a year later, and the world is different, and so am I. So, here I am at a point where I heavily rely on AI to take my notes in many different ways. Of the many ways that I use AI now, I think probably the easiest to start with is speech-to-text, [music] meaning taking my voice and turning that into **[00:30]** plain text that I can store [music] in my Obsidian vault. In this video, I'm going to show you three different levels that I've implemented this at and my workflows for each one. Here's the way that I'm currently thinking of AI's place in my notes. Nick Milo makes a distinction between note-taking, which is the passive more collecting aspect where you're just trying to write down what you heard or read, and note-making, which is connecting notes, linking them to each other, and synthesizing and **[01:00]** deriving insights from them. I think that AI naturally fits better for me, at least, in the note-taking aspect. That way, I'm kind of outsourcing a lot of the tedium of taking down notes in the first place, and not so much the synthesizing and creating new meaning from my notes. The first tier is completely free. So, no credit card information required for either the apps or the underlying LLMs that those apps are using. I think the **[01:31]** best way to start is to just use your phone. So, both iOS and Android have good alternatives for this. iOS already transcribes voice recordings on the Voice Memos app, and there is an alternative for Samsung devices as well. At least, if you have one of the later Galaxy devices, then you're going to be able to use the Voice Recorder app to do the same thing. The one that I actually use the most though is something called Spokenly. It is a free app and I'm looking at it right now. When you start **[02:03]** the app, you can look at the dictation models here. Uh if you see down here, I I'm using a free account. You can upgrade to pro if you want to use some of the frontier models, but I'm not using it for that. Instead, I'm using this uh local model. It's called Nvidia Parakeet. And so, you can choose between a lot of different ones. Apparently, there's a newer version of the one that I downloaded that is supposedly faster. There's also some Quen ones. So, these are free what's **[02:35]** called open weight local models that you can download. And you And they're the ones that do the actual transcription. So, how this works is you go into any app, let's say Obsidian, but you know, it doesn't actually have to be Obsidian. It could be like in your browser, so you could dictate an email for instance. Um I'm going to use Obsidian here, and I'm just going to create a new note. And all I have to do, and you can change what key you press. I'm pressing down the command key. And then I can just **[03:06]** start typing. You can see that there's something at the bottom there from Spokenly that notices that I'm holding this down and that I'm speaking into Obsidian. And then I'm going to let go of the right command key, and there it is. Everything I've just said is now text in Obsidian. What I like about this is it's really quick. There's no data that's being sent on the internet, so this is going to work even if you have no internet connection, and you just hold down a command key. It's pretty **[03:36]** easy to trigger it. You can also change it so that it's more of a toggle thing, so you press once to start the recording, and then you press again to stop it. But I find myself using Spokenly more for like quick sort of thing, so I just like to hold it down anyway. Now, this free tier is what I would recommend if you kind of just want to try it out and see if this will make it into your workflows. This is, I think, the most accessible one because you don't have to pay any money. What **[04:08]** it's not so good for is longer content. You know, you can do the toggle thing like I mentioned, but there is no asking AI about what you've just recorded. It was really just translating from words that you you speak to words that you've written, and that's it. Sometimes that's enough, but if it's not, then we go into tier two, which is the paid tier. In tier two, you're basically paying for the convenience layer. So, in tier two, you have these paid apps which often **[04:40]** have a free component, but I've found that it's really more realistic to think of it as paid because the free tiers are not that generous. These are tools that you might already be familiar with if you're having meetings online. So, Google Meet, for example, has Gemini that can record and transcribe what is being said. Zoom has their AI companion as well. Um Otter AI is one that's been out for a while, but the ones that I use, there are two of them, are Voice **[05:10]** Notes and Granola. Now, Granola is really good for meetings, and I really like that here in Granola, it brings in my calendar, so I've linked it to my Google Calendar, so it's aware of the meetings that come up. And whenever I start one of them, a little notification comes up in the corner that says, "Do you want to take notes?" And it's very easy to just say, "Yes, take notes." This is actually the summary that was generated from a video I watched by a YouTuber called BookBorn. My partner and **[05:41]** I are wanting to read the latest Dungeon Crawler Carl book, and we've already forgotten the previous seven books. So, I watched her recap of it, and this is the notes that Granola automatically took of it. A really cool thing here is that with Granola, you can ask questions live. So, I obviously did this already, and it's not live, but I can still kind of show you what that looks like. So, I'll say, "Uh who is Princess Donut?" **[06:11]** And I'll enter that, and in real time, or apparently after the fact, it can answer questions. So, now it's saying, "Princess Donut is Carl Sapient's cat companion, and one of the two central protagonists in the book." In meetings, I actually find this really useful if I get distracted, especially in long meetings, and I'm like, "Can you just recap the last 5 minutes because I was not paying attention?" Let me flick back to Obsidian. This is my actual Obsidian note for it. This is the video that I was watching. I created a macro. So, now **[06:45]** I just have to open up the command pane, and then um and I type Granola, and I have this link Granola one. I hit enter, and then I can choose which one it was. So, in this case, it was Dungeon Crawler Carl. So, now the Granola section is added to the bottom of the note. So, I I set all of this up in the macro. And now, that summary is here as well. So, every time I open this Obsidian note, the entire summary is also there. And **[07:15]** then, at the bottom here, uh you can also look at exactly what was being said there, and it's like nicely formatted, too. So, this is using two things. One is the Granola plugin. Uh it's called Yeah, Granola sync. It is a community plugin, and then it's also using a specific script that I wrote just to be able to link Granola transcripts and summaries to the notes to the corresponding notes in Obsidian. **[07:46]** Now, you can ask your AI to write yourself a script that works for you. If you want a copy of mine in particular, I do have it in my Patreon vault, so it is also here Granola. The other thing that I use is voice notes. Now, I tend to use voice notes actually mostly on my phone. My use case of voice notes is actually a little too personal to show because it's more for like diary style entries. So, I can't show you the interface, but you can also ask the AI there for certain **[08:17]** things um that span different notes. So, how many times have I talked about this or or when was the last time that I talked about this person? So, that kind of thing that requires multiple notes, it's pretty good at retrieving. And on Obsidian, there's also a voice note sync app that works really well. And just like you saw with Granola, I do also have a macro that runs a script for linking voice notes to my daily notes because my use case of it is more like a journal. So, I want it always embedded **[08:49]** in my daily notes. Some tradeoffs on this tier is, well, the cost and also the fact that you are losing a little bit in privacy because you are now sending those notes to other people to store on their servers. There's also the risk of that whole workflow getting interrupted if that that subscription service is no longer functional. So, you do take a little bit more risk with it. What you gain though is a lot more convenience. I find that I do voice notes in Granola actually more than I **[09:20]** use Spokenly, and I think it's just because it's it's smarter. So, I think that if you're someone who has something specific in mind to capture audio for like a talk at a conference or a video or a voice memo, and you're okay with starting and stopping it, then I think the tools in this section are for you. If you want even more, >> [laughter] >> then there's like tier three, which is the overkill, completely ridiculous level that I'm going to show you a **[09:51]** workflow for anyway, because I am that ridiculous. Let me introduce you to Plaud. You might not have noticed this on my wrist, but it is a little hardware recorder. It is something that sits on your wrist here. Uh it doesn't come with this band. This is something that I got off market, but it comes with a much uglier band, or it comes with a necklace, or a pin, and I kind of mix it up a little, um based on how obtrusive or unobtrusive I **[10:23]** want it to be. It is basically a recorder, and you can just use it as a recorder, so you can just have it as something that you wear that records your voice, and then at the end of the day, you can transfer all of those recordings to your laptop or your mobile device, where you can then use AI to transcribe it, or you can optionally send those recordings to their cloud. So, what I do here is I use this Plaud Sync plugin, which I did not create. It **[10:55]** is by Leonard Selim, but I found that there were some things that I wanted to change that I wasn't sure that everybody would really want. So, you can download his version and install it, or if you want, you can also download my version of it, which is on Nicole van der Hoeven - Plaud Sync for Obsidian. I'll put a link in the description, and it just formats it a little better for this particular use case. It works for me, you know, you I would encourage you to **[11:26]** choose the version that works best for you. So, that is the Obsidian part of it. That's how I get these transcripts into Obsidian. And when they come through, they're already nicely formatted and summarized. One cool thing that I like about Plaud that I don't get with Granola or Voice Notes is that Plaud lets you have templates. So, they're not always just processed in a certain way. Sometimes I've recorded like games, for example. And when I'm playing a role-playing game, I do not want to have like next actions and **[11:59]** agenda items the way I would in a meeting. So, I have a game one and I have a meeting template. And then I choose the more relevant template for what I'm recording. But one thing that I've actually been experimenting with is just using it for life logging. So, turning it on at the beginning of the day and letting it just record everything. Because Plaud devices, or at least this Plaud note pen, there are a few kinds of devices including like a card-shaped one. Um, but they can record for like up to **[12:31]** 23 hours or more, I think. So, I've also been experimenting with using it as a life logging tool. I really love to travel and my partner and I are really enjoying seeing new parts of Portugal. And so, with his consent, obviously, I have recorded the entire day of our conversations. And those conversations I then feed to my Open Claw agent. Now, I'm not going to get into Open Claw here, but basically it is Think of it as **[13:01]** an AI assistant with a persistent memory. So, this is something that knows you and has an extend a much more extensive memory of what you like and the things that you're interested in. And it's also agentic. So, it can do things like create notes. So, I'm going to show you just briefly how that translates in my Obsidian vault. This is my daily note for May 2nd. And you can see that I have a section here for plodd. That's because I do have um macro **[13:34]** for that as well. And it is this one, this linking plodd to the daily note. Each individual one looks like this. So that this one is I I was using a template where it's really just a transcript and no summaries. And it has these timestamps here for the raw transcript. Now, my open claw agent Iris, and if you want to know why I call her Iris, then check out this video up at the top where I talk a little bit more about that setup. Looks through all of the plodd transcript. So in this case **[14:05]** it was 15 hours of recording time or or less cuz it's 5 hours per recording. And she created my open claw assistant created a daily summary of it. So talks about where we were, the Sintra coast, and a restaurant that we went to. She basically took my voice transcripts and turned it into a summary of the day. And these are things that I now don't remember. Like do I remember that it was called the Moschino? Mm **[14:36]** no. All right. This is not without its cons. First, it has to be addressed. There are real ethical concerns here. I try not to record people, especially if they say that they don't want to be recorded. That's no questions asked. I will stop that for you. But the problem is that there are all sorts of gray areas where, you know, what if I'm in a conference and I'm speaking to someone and that person agrees to be recorded, but another person just suddenly comes up and starts talking. That third person **[15:09]** is now being recorded. So do I then have an obligation to stop and say like, "Hey, we're recording this. Is that still okay?" That can be unwieldy in a conference setting, for example, which is actually what I got it for so that I can remember who said what without actually stopping and taking notes. Um, so I'm still trying to figure out what where I am on that spectrum and you can figure that out for yourself if if you want that too. One of the benefits is that I no **[15:40]** longer have to stop at the end of the day and think, "Where did I go? What was the name of that restaurant?" Now, it just is kind of passively recorded. At the end of the day, I still have a not perfect, but way better record of what I've done than what I normally have time for. So, regardless of which of the three levels you choose or if you choose any at all, what I'm getting out of this is that [music] I think voice is now the least friction **[16:10]** capture method that exists for me because I get to just do the thing and live my life and have it passively recorded and my tasks waiting for me when I get [music] home back to my Obsidian vault. The real overkill thing here is that I still use all three levels. I use spokenly for just writing an email or dictating something into Obsidian. I use voice notes for more personal diary type entries, Granola for meetings, videos, and games, and I use Plod as my kind of daily life logger and **[16:44]** capturer of things that I didn't think to capture. The result of implementing these tools is that for me, the friction between thinking of something and writing it down is basically gone for me. And I've always been a note taker, but I still didn't take notes on everything. Now, it's still not everything, but it's pretty damn close. And that means that if I can outsource that part of it, then I get to focus on being present, staying in the moment, and if I choose to take notes, it's not **[17:16]** going to be what I heard or what happened or anything surface level like that. I can focus on the actual synthesis. So, I'm getting to that connecting, synthesizing part earlier on in the cycle. If you want any more details on any of these workflows, then drop something in the comment below so that I know that I should make a video on that.