Online Course Creator's Transcription Playbook

A 40-lesson course library is a library of locked doors. The video files are searchable only if a student remembers the exact lesson where you mentioned that one concept. Transcripts open every door at once.

BrassTranscripts converts course video lessons into searchable SRT, VTT, TXT, and JSON transcripts in 1-3 minutes per hour of footage, so students who skim and search find your material instead of bouncing. This playbook walks through the full workflow from a single lesson recording to captions, downloadable resources, SEO-ready blog posts, quiz questions, and lead magnets — one upload, six outputs.

Quick Navigation

• Online Courses Lose Students Where Search Fails
• The Caption Problem Course Platforms Don't Solve
• The Full Playbook
• Accessibility Compliance: Why Captions Aren't Optional
• SEO: Your Lessons Become Findable Pages
• AI Prompts for Course Creators
• Frequently Asked Questions

Online Courses Lose Students Where Search Fails

BrassTranscripts converts course video lessons into searchable SRT, VTT, TXT, and JSON transcripts in 1-3 minutes per hour of footage, so students who skim and search find your material instead of bouncing. Video without a transcript is a black box. Students can't scrub, can't search, can't cite a sentence in a community post without rewatching the whole module.

The cost shows up as completion rate. A student who can't find the slide on pricing strategy at minute 23 of lesson 4 doesn't watch the whole thing again. They give up and post in the Discord. You answer the same question for the fifth time. Your support load climbs while completion drops.

Transcripts fix this at the root. A searchable text version of every lesson means students self-serve. A downloadable transcript means they study on the train without burning data. A captioned video means they watch with the sound off in a coffee shop. Every one of those is a student who keeps going.

The platforms know this. Teachable, Kajabi, Skool, Thinkific, Podia, and Mighty Networks all built auto-caption features. They also all generate inconsistent captions, refuse to export them cleanly, and leave you stuck with whatever quality you got the first time.

The Caption Problem Course Platforms Don't Solve

BrassTranscripts produces clean SRT, VTT, TXT, and JSON files that work across Teachable, Skool, Kajabi, Thinkific, Podia, and Mighty Networks, while the built-in auto-caption tools on those platforms generate inconsistent output and don't export portable files. Course platform captions are good enough for a demo and not good enough for a $497 product.

Three specific failures show up across every platform we've tested:

Inconsistent technical vocabulary. Course content is dense with jargon: "lead magnet," "EBITDA," "Bayesian inference," "TypeScript generics," "kettlebell snatch." Auto-caption engines built into video platforms default to general-English models. Your industry terminology comes back as gibberish. A finance course full of "WACC" and "FCFE" gets rendered as "whack" and "F.C. fee."

No portable export. Captions generated inside Teachable or Skool live inside Teachable or Skool. If you later move to Kajabi or republish a lesson on YouTube, you start over. There's no SRT download button. You can't feed those captions into an AI tool. You can't post the transcript on your blog. The work is trapped.

Limited multi-speaker handling. Interview-style lessons with a guest, panel-format modules, and Q&A sessions don't get speaker labels in most platform captions. A student reading the transcript can't tell who said what. The format breaks down exactly when you need it most.

The fix is to run your master audio through a dedicated transcription pass before you ever upload to the course platform. You get clean files in every format you'll need across the lesson's entire life — captions, transcripts, blog posts, AI inputs.

The Full Playbook

Here's the workflow from raw lesson recording to seven assets. Each step compounds on the previous one. Do step 2 once, get steps 3 through 7 at almost no extra cost.

Step 1: Record the Lesson With Audio Quality That Transcribes Cleanly

Video quality matters less than people think for online courses. Audio quality matters more. A great-looking lesson with hollow room audio sounds amateur; a plain talking-head shot with crisp mic audio sounds expert. Transcripts amplify this — clean audio produces clean transcripts; muddy audio produces transcripts full of [inaudible] markers.

The practical bar: a USB condenser or dynamic mic ($60-150), recorded in a small room with soft furnishings, at 48kHz/16-bit or higher. Avoid the laptop mic. Avoid AirPods. Headset mics from gaming brands work surprisingly well. For deeper audio guidance, see audio quality secrets for perfect transcription.

Record video and audio separately if your editing tool supports it. The audio track is what you'll send to transcription. The video gets edited around it.

Step 2: Upload the Lesson to BrassTranscripts and Get Four Formats

Drop the audio or video file at brasstranscripts.com. Files up to 450MB go through in one shot. Processing runs at 1-3 minutes per hour of audio, so a 45-minute lesson is ready in about two minutes. You get four outputs:

• TXT — clean readable transcript, no timestamps, ready to paste anywhere
• SRT — caption file with timestamps, the universal subtitle format
• VTT — WebVTT, used by HTML5 video players and some course platforms
• JSON — full structured data with word-level timestamps and speaker labels

Pricing is flat per file: $2.50 for lessons under 15 minutes, $6.00 for lessons 16 to 120 minutes. No subscription. No per-minute meter. A 40-lesson course with most modules in the 16-60 minute range runs about $240 for the entire library. Compare that to the developer time you'd spend hand-correcting platform captions for one course.

For the deeper version of which format does what, see how to choose the right transcript format: TXT, SRT, VTT, or JSON.

Step 3: Upload the SRT to Your Course Platform's Caption Track

Every major course platform accepts SRT upload on the lesson video player:

• Teachable: Lesson editor → Video → Captions → Upload SRT
• Thinkific: Course Builder → Video lesson → Subtitles → Add subtitle file
• Kajabi: Product → Lesson → Video settings → Add captions
• Skool / Mighty Networks: Use an external host (Vimeo, Wistia) that supports SRT and embed
• Podia: Lesson editor → Video → Captions

Set the caption language to match the spoken language. The result: a "CC" button on every lesson, accessibility compliance moved from a worry to a checkbox.

Step 4: Publish the TXT as a Downloadable Lesson Transcript

The TXT file becomes a student resource. Wrap it in a simple branded PDF with the lesson title, your logo, a one-paragraph summary, and the full transcript. Upload to the lesson page as a download.

This single asset solves three student behaviors:

• The note-taker — gets a complete transcript instead of pausing every 30 seconds
• The reviewer — scans the transcript in two minutes six months later
• The accessibility user — uses screen readers, prints to read on paper, or translates to another language

The workflow is well-established in education and translates directly from classroom to course platform — see lecture transcription: a student's study guide for the study-side perspective.

Step 5: Repurpose Transcripts Into Blog Posts (the SEO Play)

This step is where the math changes. A 45-minute lesson produces 6,000-8,000 words of raw transcript. Edited down to remove ums and tangents, that's a 3,000-5,000 word blog post — long-form content that ranks. One transcript pass, one editing afternoon, one blog post per lesson.

The workflow:

1. Open the TXT in your editor
2. Strip filler words with find-and-replace ("um," "uh," "you know," "right?")
3. Insert H2 headings at natural section breaks
4. Add a 2-3 paragraph intro and a CTA at the end
5. Pull one quote into a pull-quote block
6. Publish on your blog with a link back to the course

A 40-lesson course becomes 40 blog posts. Each post targets long-tail keywords your competitors don't write about. The blog posts feed organic traffic into the course funnel, and Google finally sees content that used to live behind a paywall. Deeper mechanics are in the complete video transcription guide.

Step 6: Generate Quiz Questions and Review Summaries From the JSON

The JSON output is structured data — segments, word-level timestamps, speaker labels. Feed it to ChatGPT or Claude with a prompt and you get quiz questions tied to specific lesson timestamps, review summaries that quote the lesson with citations, and concept maps that show how lessons connect.

This matters for two reasons. Certification programs need defensible quiz questions tied to source material — a timestamp citation ("see lesson 4, 12:35-14:10") makes the question auditable. Review summaries posted as lesson recaps boost retention. Prompts for both are in the section below.

Step 7: Build a Lesson Search Index for Your Course

Some course platforms support custom search (Skool's community search, Kajabi's site search). Others don't, and you end up with a course where students can't search across lessons at all. The JSON transcripts make this fixable.

Two approaches work:

• Simple: paste every lesson's TXT into a single search-engine-indexable page on your course site. A student searches the page in-browser with Ctrl+F. Crude, but it works on day one.
• Better: feed all JSON files into a vector database (Pinecone, Weaviate, Supabase pgvector) and put a chat interface on top. Students ask "where did Sarah cover positioning?" and get a direct timestamp link. This is the level competing course platforms charge $50/month extra for.

For a 40-lesson course, the entire transcript corpus is small enough to fit inside an LLM context window. A bare-bones implementation is a weekend project.

Accessibility Compliance: Why Captions Aren't Optional

The Department of Justice has applied ADA Title III to commercial websites and online services, and the W3C WCAG 2.1 Level AA guidelines require synchronized captions for prerecorded video. Online course platforms count as places of public accommodation under the prevailing Title III interpretation, which means caption requirements apply to your course content.

The specific guideline is WCAG 2.1 Success Criterion 1.2.2 (Captions, Prerecorded), Level A — captions are required for all prerecorded audio content in synchronized media. Level AA adds 1.2.5 (Audio Description, Prerecorded) for video-only content. Courses recorded as screen-shares with voiceover trigger both.

Beyond the legal frame, the audience math is large. The CDC reports that approximately 15% of American adults (37.5 million people) have some hearing trouble. ESL learners prefer captions even when not required, and captioned content earns more shares and time-on-page across every platform that has measured it.

For the deeper compliance walk-through with full citations and platform-by-platform requirements, see ADA compliance for transcription and accessibility requirements.

The practical takeaway: ship captions on every lesson from day one. Adding them later means rebuilding the lesson library. Adding them upfront costs $2.50 to $6.00 per lesson and takes minutes.

SEO: Your Lessons Become Findable Pages

Course content locked behind a paywall is invisible to Google. Transcripts published as blog posts make that content indexable, and a single 45-minute lesson typically contains enough material to target dozens of long-tail keyword phrases that competitors don't write about. One lesson becomes one indexable page, scaling linearly with your course library.

The strategic shift: stop writing "marketing blog posts" about your topic. Start publishing edited lesson transcripts. Marketing blog posts are written for SEO and feel hollow. Lesson transcripts are written for paying students and read as expert material — which is exactly what Google's helpful-content updates reward.

Two SEO mechanics make this work. First, long-tail keyword density: a finance lesson on cash-flow modeling naturally contains phrases like "DCF terminal value assumptions" and "working capital normalization" that a marketing copywriter would never include. Your transcript does, because you actually taught it. Second, internal link gravity: forty lesson-derived posts on a single topic interlink naturally, and Google reads the cluster as topical authority.

The catch: publish lesson transcripts as a complement to the paid course, not a replacement. Strip out the screen-share references and in-course exercises. Keep the teaching content. Keep the CTAs pointing to the course.

AI Prompts for Course Creators

These three prompts cover the highest-value AI workflows for course transcripts: turning lessons into blog posts, generating quiz questions, and pulling FAQs for sales pages. Paste each into ChatGPT, Claude, or your AI tool of choice; replace the bracketed placeholders.

Lesson-to-Blog Converter

The Prompt

You are an editor converting a course lesson transcript into a publishable blog post.

INPUT: The full TXT transcript of a course lesson is pasted below.

LESSON TITLE: [Lesson title]
COURSE NAME: [Course name]
TARGET KEYWORD: [Primary SEO keyword for this post]
AUDIENCE: [Who the course is for]

YOUR TASK:
1. Remove all filler words (um, uh, you know, right?, like).
2. Remove all in-course references (e.g., "as we saw in lesson 4", "your workbook says", "the homework for this week").
3. Restructure into a blog post with:
   - A 2-paragraph intro that hooks the reader on the target keyword
   - 4-6 H2 sections with title case headings
   - One H3 subsection per H2 where it adds clarity
   - A 3-sentence conclusion
4. Preserve the teacher's voice. Do not rewrite for "marketing tone." Keep the practitioner voice intact.
5. Insert one pull-quote (a single strong sentence the teacher said) about 60% of the way through.
6. End with a CTA paragraph pointing to the course (CTA text: "[Your CTA]").

CONSTRAINTS:
- No marketing fluff
- No banned SEO words ("ultimate guide", "definitive", "everything you need to know")
- Target word count: 1500-2500 words

TRANSCRIPT:
[Paste TXT transcript here]

---
Prompt by BrassTranscripts (brasstranscripts.com) – Professional AI transcription with high-quality results.
---

Quiz-Question Generator From Transcript

The Prompt

You are a curriculum designer generating quiz questions for an online course lesson.

INPUT: The JSON transcript of a course lesson is pasted below, with word-level timestamps and segments.

LESSON TITLE: [Lesson title]
LEARNING OBJECTIVES: [List 3-5 learning objectives for this lesson]
DIFFICULTY: [beginner / intermediate / advanced]

YOUR TASK:
Generate 8 quiz questions tied to specific lesson timestamps:
- 4 multiple-choice (4 options each, one correct)
- 2 true/false
- 2 short-answer (1-2 sentence expected response)

For each question:
- Include the correct answer
- Include a "source timestamp" referencing the JSON segment where the answer is taught (format: MM:SS-MM:SS)
- Include a brief explanation of why the answer is correct, quoting the lesson where helpful

QUESTION CRITERIA:
- Test understanding, not memorization of specific phrasing
- Each question must be answerable from the lesson content (no outside knowledge required)
- Cover all listed learning objectives across the 8 questions
- Vary difficulty — at least 2 questions should require synthesis across multiple segments

OUTPUT FORMAT: Markdown table with columns: Question | Type | Correct Answer | Source Timestamp | Explanation

TRANSCRIPT JSON:
[Paste JSON transcript here]

---
Prompt by BrassTranscripts (brasstranscripts.com) – Professional AI transcription with high-quality results.
---

FAQ Extractor for Sales Pages

The Prompt

You are a sales-page copywriter extracting frequently asked questions from course content.

INPUT: TXT transcripts from multiple course lessons are pasted below. Each lesson is separated by a clear delimiter.

COURSE NAME: [Course name]
AUDIENCE: [Who the course is for]
PRICE POINT: [Course price]

YOUR TASK:
Identify the 8 most common implicit questions the teacher answers across the lessons. Convert each into:
- A clear, customer-voiced question (as a prospect would phrase it)
- A 2-3 sentence answer that pulls the teacher's actual reasoning from the transcripts

CRITERIA FOR INCLUSION:
- The question must be one a prospect would have before buying (objections, prerequisites, outcomes, time required, tools needed)
- The answer must come from the teacher's actual content, not invented
- Prefer questions answered across multiple lessons (signals the question is core)

OUTPUT FORMAT:
## Frequently Asked Questions

### [Question 1 in customer voice]
[Answer pulling from teacher's content]

### [Question 2 in customer voice]
[Answer pulling from teacher's content]

...

Generate 8 Q&A pairs total. Do not invent questions the teacher does not answer in the transcripts.

TRANSCRIPTS:
[Paste TXT transcripts here, separated by "--- LESSON BREAK ---"]

---
Prompt by BrassTranscripts (brasstranscripts.com) – Professional AI transcription with high-quality results.
---

Frequently Asked Questions

Do Teachable, Kajabi, and Thinkific generate captions automatically?

Teachable, Kajabi, and Thinkific offer auto-caption features, but the output quality is inconsistent for technical vocabulary, accented speakers, and multi-speaker recordings. None of them export clean SRT files for reuse on other platforms or VTT for embedded players, which is why most serious course creators run an external transcription pass.

What transcript format should I upload to my course platform?

Upload SRT for caption tracks on most course platforms because SRT is the most widely supported subtitle format across Teachable, Thinkific, Kajabi, Skool, Podia, and Mighty Networks. Use VTT for HTML5 video players that require WebVTT, TXT for downloadable lesson transcripts, and JSON when feeding data into AI tools for quiz generation or search indexing.

Are captions legally required for online courses?

The Department of Justice has applied ADA Title III to commercial websites and online services, and the W3C WCAG 2.1 Level AA guidelines require synchronized captions for prerecorded video. Captions reduce legal exposure and broaden the addressable market to deaf and hard-of-hearing students, ESL learners, and viewers in sound-sensitive environments.

How much does it cost to transcribe a course library?

BrassTranscripts charges $2.50 per lesson under 15 minutes and $6.00 per lesson between 16 and 120 minutes. A 40-lesson course with mostly short modules runs around $100 to $240 total. Processing takes 1-3 minutes per hour of audio, so a full course library is usually ready the same day.

Can transcripts help my course rank in Google?

Publishing lesson transcripts as blog posts gives Google indexable text for content that would otherwise live behind a paywall as video only. One 45-minute lesson typically yields 6,000 to 8,000 words of transcript, which converts into a long-form blog post that targets dozens of long-tail keyword phrases per lesson.

What can I do with the JSON transcript output?

The JSON output from BrassTranscripts includes word-level timestamps, speaker labels, and segment data. Course creators feed JSON into AI tools like ChatGPT or Claude to generate quiz questions tied to specific lesson timestamps, build lesson search indexes that jump students to the exact second a concept is mentioned, and produce review summaries that pull quotes with citations.