More project photos (3)
The problem
Podcast episodes routinely run one to three hours, yet most players offer nothing better than a scrub bar — listeners cannot jump to the interview segment or skip the sponsor read. Creators know chapters fix this, but hand-writing timestamps for every episode is tedious enough that most skip it. Speech recognition changed the economics: Whisper produces timestamped transcripts good enough to work from, and topic shifts in the transcript reveal natural chapter boundaries. This project wires that into a complete pipeline — transcribe, segment, title, review on a waveform, export — so a creator uploads an episode and gets ready-to-paste chapters. The listing is careful about what is automated and what is not: transcription quality depends on the audio, and chapter boundaries are heuristic suggestions the user reviews, not perfect cuts.
Uploads episodes, reviews chapter boundaries and edits titles.
Follow the working sequence
How it works
- The user uploads an MP3/WAV episode; FFmpeg normalizes it to the 16 kHz mono format Whisper expects.
- faster-whisper runs the selected pre-trained Whisper model and returns segments with start/end timestamps.
- A segmentation stage combines long silences with topic-shift signals in the transcript to propose chapter boundaries near the target length.
- Each proposed chapter receives an extractive title from its most distinctive phrases.
- The web UI renders the waveform with chapter regions; the user drags boundaries, splits, merges and retitles as needed.
- Export writes YouTube-format chapter text, MP3 chapter tags and JSON from the reviewed chapter list.
Tech stack:
- Python 3.10, faster-whisper (CTranslate2 runtime)
- OpenAI Whisper pre-trained weights (open-source; downloaded once)
- FFmpeg (audio normalization and chapter tagging)
- Flask backend, HTML/CSS/JS waveform editor frontend
| Parameter | Value |
|---|---|
| Transcription model | Whisper pre-trained (tiny/base/small/medium/large selectable) |
| Training required | None — pre-trained weights are used as-is |
| Input | MP3/WAV podcast audio, any length |
| Output | Timestamped transcript, titled chapters, YouTube/MP3/JSON exports |
| Chapter titles | Extractive (from segment phrases), user-editable |
| Runtime | CPU supported; GPU strongly recommended for large models |
| Transcription quality | Depends on audio quality, accents, crosstalk and music — no WER claimed |
Project features
- [Whisper transcription] Pre-trained Whisper models (tiny through large, selectable) transcribe any episode with segment-level timestamps; no model training is required.
- [Automatic chapter segmentation] Long silences and topic shifts in the transcript are combined to propose chapter boundaries, with a configurable target chapter length.
- [Extractive chapter titles] Each proposed chapter gets a title drawn from its most distinctive phrases — transparent and inspectable, not a black-box generator.
- [Interactive waveform editor] The episode renders as a chapter-colored waveform; boundaries can be dragged, chapters split, merged and retitled before export.
- [Multi-format export] One click produces YouTube description chapters, MP3 ID3 chapter tags and a JSON chapter file for other players.
- [Batch queue] Drop in several episodes and process them sequentially, with per-episode status and logs.
- [Configurable pipeline] Model size, language hint, silence thresholds and target chapter length are all adjustable from the UI.
What is included
- Complete source code (transcription pipeline, segmentation, web UI, exporters)
- Setup guide (Python environment, Whisper weight download, FFmpeg, running the UI)
- Project report PDF (background, Whisper overview, segmentation method, evaluation procedure)
- PPT presentation for final review
- Viva Q&A preparation document (speech recognition, transformers, timestamp alignment, segmentation heuristics)
- Demo guidance with a sample episode workflow
Limitations & prerequisites
- Transcription errors rise with heavy accents, crosstalk, background music and poor microphones — the UI is built for reviewing, not blind trust.
- Chapter boundaries are heuristic suggestions; topic shifts in casual conversation do not always match, so human review matters.
- Large Whisper models need significant RAM/VRAM and time on CPU; the tiny/base models are provided as fast fallbacks.
- Very long episodes (3+ hours) need patience and disk space for intermediate files.
- The pipeline does not separate overlapping speakers; it chapters topics, not diarized speakers.
Frequently Asked Questions
Which model is used?
OpenAI's Whisper, using its published pre-trained weights (tiny through large). No training is done in this project — it is applied transcription, not model development.
Does it need training data?
No. Whisper arrives pre-trained; the project's own contribution is the segmentation, titling, waveform review UI and exporters.
Is the transcription accuracy guaranteed?
No. Quality depends on the recording — clear studio speech transcribes well, while noisy, accented or music-heavy audio degrades. The report documents this openly.
Do I need a GPU?
Not strictly — CPU works with the smaller models, but medium/large models are much faster on a CUDA GPU. Expected runtimes for each model size are documented.
Which languages are supported?
Whisper is multilingual; the language can be hinted or auto-detected. Chapter titling works best on English and should be reviewed for other languages.
Is this project suitable for a final-year project?
Yes — for Computer Science, AI/ML and Data Science programs. It demonstrates applied speech AI, audio processing, heuristic segmentation and full-stack product thinking. Suitable for B.E./B.Tech final-year projects in Computer Science, AI/ML and Data Science.
Components & software requirements
- Python 3.10, faster-whisper (CTranslate2 runtime)
- OpenAI Whisper pre-trained weights (open-source; downloaded once)
- FFmpeg (audio normalization and chapter tagging)
- Flask backend, HTML/CSS/JS waveform editor frontend
Delivery information
Built-to-order project. Delivery timeline is shared after order confirmation based on current queue.
Support terms
Complete documentation, setup guide, and viva preparation included. Support for setup and explanation provided.
Download free synopsis (PDF)
Enter your name and WhatsApp number — the synopsis PDF downloads immediately. We may send occasional project updates on WhatsApp; no spam.