Milestone map
Milestone map
2 milestones
Do the phase's lessons in public
1–1 weeks (about 10 hrs/week, est)
Work through the 17 lessons of Phase 6: Speech & Audio from AI Engineering from Scratch (free) and keep the speech and audio lessons' notebooks (spectrograms, speech recognition, Whisper, text to speech, voice pipelines) in one public repo as you go. About 10 hours (est).
Proof required
Link your public repo with the speech and audio lessons' notebooks (spectrograms, speech recognition, Whisper, text to speech, voice pipelines), and show your self-check (quizzes passed, notebooks run, tests passing where they exist).
What gets checked
- Each lesson's notebook runs.
- Speech recognition is measured with word error rate.
- Audio examples are included in the repo or linked.
Common mistakes
- Copying the repo's solutions: verifiers read your history and ask about your code.
- Skipping the self-check, then asking for review too early.
Powstik Guide
A public repo with the speech and audio lessons' notebooks (spectrograms, speech recognition, Whisper, text to speech, voice pipelines).
Steps
- Make a public repo for this phase.
- Do one lesson at a time and commit after each.
- Self-check: Run each lesson's code; it must work on your machine as the lesson shows. Run each lesson's notebook top to bottom; its outputs must match the lesson.
- Write a short README: what is here and how to run it.
- Ask three people who know the topic to check it.
Template
README Phase 6: Speech & Audio (AI Engineering from Scratch) What is here: How to run it: Self-check result: What was hardest:
What gets sent back
- One big upload with no history.
- Self-check not shown.
- Code that does not run from the README.
Resources
Foundationstart here
Depthgo deeper
What a verifier looks for
- Each lesson's notebook runs.
- Speech recognition is measured with word error rate.
- Audio examples are included in the repo or linked.
- Ask about one choice in the code; the answer should be clear.
You'll sign in first, then come straight back here.
Build a voice pipeline
2–4 weeks (6–10 hrs/week)
Use what you learned. Make a voice pipeline (speech in → text → reply → speech out) that works on recordings of at least 3 different voices, with word error rate measured. It must run, be evaluated and be documented.
Proof required
Link a voice pipeline (speech in → text → reply → speech out) that works on recordings of at least 3 different voices, with word error rate measured, with a README that says how to run it and what the evaluation found.
What gets checked
- It runs from the README.
- Word error rate is measured per voice.
- Latency from speech in to speech out is measured.
Common mistakes
- A demo with no evaluation.
- Numbers without the data or code that produced them.
Powstik Guide
A voice pipeline (speech in → text → reply → speech out) that works on recordings of at least 3 different voices, with word error rate measured.
Steps
- Write one line: what it does and for whom.
- Build the smallest version that works.
- Write the evaluation first, then improve against it.
- Write the README with the results.
- Ask three people to check it.
Template
README What it does: How to run it: How it was evaluated: Results: Limits:
What gets sent back
- Does not run from the README.
- No evaluation.
- Claims more than the evaluation shows.
Resources
Foundationstart here
What a verifier looks for
- It runs from the README.
- Word error rate is measured per voice.
- Latency from speech in to speech out is measured.
- Run it, or watch it run: it must work, be evaluated and be documented.
- Check that the work is the person's own.
You'll sign in first, then come straight back here.