Transistor.fm: AI Transcripts for Podcasts  🔗 **Source:** [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > [!note]+ Document Notes > tags: [Podcasting, AI-Transcription, Workflow] > author: "[MISSING]" > source: YouTube > reference: [https://www.youtube.com/watch?v=qTjRtUlJeTw] > date created: {{date:YYYY-MM-DD}} {{time:HH:mm}} > > # 🏷️ Title: Transistor.fm: AI Transcripts for Podcasts > <!-- deep-read:sections start --> ## 🧰 Action Inbox - [H1] Publish quick-start for enabling Transistor transcripts — context/https://www.youtube.com/watch?v=qTjRtUlJeTw&t=2 #articleseed - [H2] Document in-app editing workflow and speaker labeling — context/https://www.youtube.com/watch?v=qTjRtUlJeTw&t=95 #xref - [H3] Create checklist for transcript QA fixes — context/https://www.youtube.com/watch?v=qTjRtUlJeTw&t=153 #articleseed - [H4] Draft SOP for team-based transcript roles — context/https://www.youtube.com/watch?v=qTjRtUlJeTw&t=182 #xref - [H5] Compare Transistor vs. Descript/Riverside handoffs — context/https://www.youtube.com/watch?v=qTjRtUlJeTw&t=28 #articleseed - [H6] Add pricing note with audit date — context/https://www.youtube.com/watch?v=qTjRtUlJeTw&t=216 #addendum - [H7] Propose WordPress plugin docs section on transcripts — context/https://www.youtube.com/watch?v=qTjRtUlJeTw #share - [H8] Write aphorism: “Automation drafts; editors finish.” — context/https://www.youtube.com/watch?v=qTjRtUlJeTw #aphorisms ## 🧭 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) ### 🔗 References - Video: Transistor.fm — AI Transcripts for Podcasts — https://www.youtube.com/watch?v=qTjRtUlJeTw - Transistor.fm — https://transistor.fm/ - Transistor podcast transcripts feature — https://transistor.fm/podcast-transcripts/ *(If page changes, see video reference.)* - Descript — https://www.descript.com/ - Riverside — https://riverside.fm/ - Crew Collective Podcast — https://crewcollectivepodcast.com/ ## 📝 Summary - The video demonstrates Transistor’s AI transcript add-on. - It automates transcript creation inside the hosting platform. - Creators, editors, and podcast teams benefit from faster edits. - Risks include transcription errors and unclear pricing changes. - Bottom line: automation saves time, but human QA still matters. ## 📖 Main Content - Transistor generates transcripts directly for each episode. - Users can upload a file or use automated transcription. - The interface supports speaker labeling and paragraph edits. - Teams can share transcript duties using built-in roles. - The host states tiered pricing for transcript hours. - Evidence remains anecdotal; no WER metrics appear. - Compare with Descript and Riverside for alternatives. - Keep an audit trail when changing speaker names. - Publish a QA checklist to reduce copy errors. ### 🗒️ Notes #### ⏱️ [00:02](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=2) Feature overview ### [H1] Transistor provides automated AI transcripts > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) by an unnamed author via [[YouTube]]. Published [MISSING]. Accessed {{date:YYYY-MM-DD}}. > “Transistors transcription features uh powered by AI but with some really nice fine touches” Automated creation lowers manual effort and speeds publishing. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=2) #articleseed ID:rw_vid_002 #### ⏱️ [00:28](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=28) Setup and automation ### [H2] Transcripts run inside Transistor, not third-party > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > “having Transistor to just do the transcriptions… not having to copy, paste, and upload files” In-app processing reduces tool switching and errors. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=28) #xref ID:rw_vid_028 #### ⏱️ [01:22](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=82) Auto vs. manual ### [H3] Episodes can use auto or uploaded transcripts > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > “You can upload your own file… this episode was auto transcribed with AI.” Dual paths support different production workflows. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=82) #articleseed ID:rw_vid_082 #### ⏱️ [01:35](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=95) Speaker labeling ### [H4] Interface supports speaker identification and edits > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > “it did a great job voice identifying… Click the speaker name to edit.” Fast relabeling strengthens clarity and search. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=95) #xref ID:rw_vid_095 #### ⏱️ [02:04](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=124) Editing tools ### [H5] Users edit words, split paragraphs, and delete sections > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > “I can click in to edit a word. I can split the paragraphs.” Granular edits enable quick fixes without exports. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=124) #articleseed ID:rw_vid_124 #### ⏱️ [03:00](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=180) Team roles ### [H6] Teams can manage transcripts collaboratively > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > “you can add teams to a podcast… they could log into Transistor and edit” Role-based access supports distributed production work. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=180) #xref ID:rw_vid_180 #### ⏱️ [03:16](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=196) Pricing tiers ### [H7] Host states per-hour pricing by monthly tiers > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > “$5 a month extra for up to 5 hours… 20 for 20 hours, and 40 for 40 hours.” Treat as indicative; verify on Transistor’s site. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=196) #addendum ID:rw_vid_196 #### ⏱️ [03:36](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=216) Affiliate note ### [H8] Video includes affiliate link disclosure > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > “It’s an affiliate link. Helps the channel. You don’t have to use it.” Consider potential bias when evaluating claims. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw&t=216) #xref ID:rw_vid_216 ## 📚 References - Transistor feature overview — https://transistor.fm/podcast-transcripts/ - Transistor pricing page — https://transistor.fm/pricing/ *(If pricing differs from video, prefer site data.)* - Descript transcripts — https://www.descript.com/transcription - Riverside transcripts — https://riverside.fm/features/transcriptions - Crew Collective Podcast — https://crewcollectivepodcast.com/ ## 🗣️ Paraphrase - Transistor automates transcripts and enables quick in-app edits. - Teams can collaborate on labeling and corrections with minimal overhead. ## 📰 Newsletter - Transistor now ships AI transcripts inside the host UI. - You can fix names, split paragraphs, and assign speakers fast. - Teams can share transcript work without exporting files. - Pricing appears tiered; confirm on Transistor’s site. - Link roundup: Transistor, Descript, and Riverside comparisons. ## ✅ Action Items #### DevRel - Record a 3-minute “edit a transcript” demo with captions. - Publish a guide on speaker-label best practices for hosts. - Share a prompt-free workflow since edits happen in-app. - Track time saved versus manual export flows. #### OSPO - Document data handling and retention for transcript text. - Review consent policies for guest transcripts and distribution. - Add guidance on affiliate disclosures for staff content. - Create redaction SOPs for sensitive statements. #### Community - Host an office hour on transcript QA workflows. - Share a checklist for labeling and paragraph fixes. - Collect feedback on accuracy and missing features. - Provide examples for multilingual shows. #### WordPress - Documentation: add how-to for embedding transcripts on posts. - Plugins: inventory transcript-friendly themes and blocks. - Accessibility: recommend captions plus downloadable text files. - Support: create FAQ on transcript corrections. #### The WPCC - Fund microgrants for accessibility-first podcast templates. - Sponsor workshops on transcript editorial standards. - Publish impact reports showing accessibility gains. ### FAIR - Define provenance fields for transcript edit histories. - Draft policies for retention, portability, and export formats. - Ship a governance scorecard for transcription vendors. *Tag Coverage — #articleseed: 3 • #xref: 3 • #share: 1 • #bmf: 0 • #addendum: 1 • #aphorisms: 1* ## 🧪 Hypotheses - In-app transcripts reduce production time by at least 20%. - Speaker labeling improves search and SEO click depth. - A QA checklist reduces post-publication fixes by 50%. - Team roles increase throughput without harming quality. ### 📊 Methodology - Run time-and-motion studies across five episodes. - Measure error rates before and after checklist adoption. - Track SEO metrics for posts with clean speaker labels. - Survey editors on perceived workload and satisfaction. ### ❓ Questions - What is the measured word error rate for Transistor? - How does pricing scale for larger catalogs? - Can users export edit histories for compliance? - Does the API expose transcript and speaker data? ### 🔮 Future Research - Benchmark Transistor against Descript and Riverside. - Test multilingual accuracy and right-to-left scripts. - Evaluate diarization quality for group panels. ### 💡 Implications - Accessibility improves when transcripts ship by default. - Editorial standards remain essential despite automation. - Teams should plan consent, retention, and provenance. ### 🧠 Learning Styles Assessment - Reading level estimate: Flesch-Kincaid Grade ≈ 9–10. #### 👀 Visual learners - Present: UI walkthrough in video. - Improve: screenshots of edit dialogs and labels. #### 🔊 Aural learners - Present: narrated demo. - Improve: 2–3 minute audio summary with highlights. #### 🗣️ Verbal learners - Present: concrete sentences and steps. - Improve: glossary for diarization and WER. #### 🧩 Kinesthetic learners - Present: hands-on edits in Transistor. - Improve: step-by-step lab checklist. #### 🧠 Logical learners - Present: clear sequences and outcomes. - Improve: decision tree for tool choice. #### 🧑‍🤝‍🧑 Social learners - Present: team roles example. - Improve: peer review rubric for QA. #### 🧘‍♀️ Solitary learners - Present: self-paced edits. - Improve: downloadable transcript samples. #### 🌿 Naturalistic learners - Present: none. - Improve: sustainability note on compute costs. ### 📊 Accessibility Assessment #### Screen reader semantics - Present: captions available on YouTube. - Improve: downloadable transcripts and alt text on images. #### Keyboard navigation - Present: standard player controls. - Improve: add chapter markers for skips. #### Contrast and readability - Present: not assessed. - Improve: ensure 4.5:1 for any shared UI. #### Text resizing and reflow - Present: responsive player. - Improve: provide large-text transcript pages. #### Media and captions - Present: auto-captions exist. - Improve: human-edited captions for accuracy. #### Forms and interactions - Present: basic links only. - Improve: clear consent and privacy notices. #### Standards reference - Map improvements to WCAG 2.2 AA criteria. ### 🌍 DEIB Assessment #### Cultural breadth - Present: English-only example. - Improve: multilingual demos and locales. #### Inclusive language - Present: neutral tone. - Improve: define jargon plainly. #### Neurodiversity support - Present: chunked walkthrough. - Improve: consistent headings and icons. #### Representation - Present: limited on-screen personas. - Improve: broader voices and roles. #### Psychological safety - Present: respectful instructions. - Improve: clear moderation guidance for comments. ### ✅ Logical Validity #### Strong reasoning - Automation saves time by removing copy steps ([H2]). - Editing tools address inevitable AI errors ([H5]). - Team roles improve throughput with accountability ([H6]). #### Fallacies - Pricing may imply stability; markets change ([H7]). - Fix: always verify with official pricing pages. #### Tone of voice - Friendly and promotional; includes affiliate disclosure ([H8]). ### 🛡️ DARVO Analysis - Deny: none apparent. - Attack: none apparent. - Reverse victim/offender: none apparent. - The rhetoric remains product-forward and supportive. ## 🎯 Conclusion - Transistor’s AI transcripts streamline podcast production. - Editors still need QA, labeling, and consent steps. - Verify pricing and retention policies regularly. - Share repeatable workflows across teams and communities. ## 🧱 Refactor Appendix (append after 🎯 Conclusion) ### 📚 Literature Split Pad - Purpose: Stage Konik CEE units that become per-source Literature notes. - Split level: **H3**. - Include verbatim quotes under 25 words and a [view highlight] link. ### [H1] Transistor provides automated AI transcripts > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) by an unnamed author via [[YouTube]]. Published [MISSING]. Accessed {{date:YYYY-MM-DD}}. > “Transistors transcription features uh powered by AI but with some really nice fine touches” Two lines: context and use. Cite data if relevant. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw#H1) #xref #articleseed ID:rw_[MISSING] ### [H2] Transcripts run inside Transistor, not third-party > [!quote] [Transistor.fm: AI Transcripts for Podcasts](https://www.youtube.com/watch?v=qTjRtUlJeTw) > “having Transistor to just do the transcriptions… not having to copy, paste, and upload files” Two lines: context and use. [view highlight](https://www.youtube.com/watch?v=qTjRtUlJeTw#H2) #xref ID:rw_[MISSING] --- ### 🔖 Atomic Split Pad (Zettel Candidates) - Purpose: Promote reusable claims into Atomic notes. - Split level: **H4**. - Each item: one idea, present tense, specific, testable. #### [Z1] In-app transcripts reduce tool switching - Measure time saved versus export pipelines. - Links: [[Transcripts run inside Transistor, not third-party]] [[Users edit words, split paragraphs]] #### [Z2] Speaker labeling improves search and clarity - Track SEO and user navigation gains. - Links: [[Interface supports speaker identification and edits]] --- ### ✏️ Fleeting Queue (Extract by Selection) - Purpose: Capture quick actions or sparks for Fleeting notes. - Extraction: **Select one line** → “Refactor: Extract selection to new note (content only).” - Keep each line short and self-contained. - [F1] Draft transcript QA checklist — context/https://www.youtube.com/watch?v=qTjRtUlJeTw #fleeting #inbox - [F2] Record 3-minute editing demo — context/https://www.youtube.com/watch?v=qTjRtUlJeTw #fleeting #inbox - [F3] Verify pricing and note audit date — context/https://www.youtube.com/watch?v=qTjRtUlJeTw #fleeting #inbox <!-- deep-read:sections end --> [[3. Resources/Readwise/Full Document Contents/Articles/Transistor.fm AI Transcripts for Podcasts|Full Document Contents]] ## Highlights - Hey, it's Matt from the podcast setup.com. I wanted to show you Transistor's transcription features, powered by AI, but with some really nice fine touches that I think make it a real treat to use. I would enable it if you have a Transistor account to start using transcripts. A fantastic advantage of transcripts, of course, is that it’s automated; it’s done for you right here in Transistor. ([View Highlight](https://read.readwise.io/read/01k81yg883e1ysma027tx1cxwk))