# jianfch/stable-ts

Transcription, forced alignment, and audio indexing with OpenAI's Whisper

Repository: https://github.com/jianfch/stable-ts
Canonical: https://ross.abutalabs.com/products/stable-ts
Language: Python
License: MIT
License Family: permissive
Archived: true
Last push: 2026-05-30T18:41:03+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 85, release rhythm 8, longevity 100
- inputs: {"age_days": 1434, "days_push": 95, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2281, forks 243 (observed 2026-08-28T04:06:33.826223+00:00)

## What it is
A Python library that modifies OpenAI's Whisper to produce more reliable timestamps, adding transcription, forced alignment, and audio indexing capabilities. It also ships a CLI for generating subtitle files like SRT and VTT from audio.

## Use cases
- transcribe audio files with accurate word-level timestamps
- generate srt or vtt subtitles from mp3 recordings
- align existing transcripts to audio with forced alignment
- locate when specific words occur in an audio file
- regroup and refine whisper subtitle segments
- suppress silence gaps in transcription output

## When to avoid
- you need actively maintained software - development is indefinitely paused
- you need non-Whisper ASR engines as your primary workflow
- you need real-time streaming transcription

## Facets
- artifact type: library
- maturity: maintenance
- function: speech-recognition, audio-processing, nlp, cli
- domain: speech-processing, developer-tools
- platform: python, cli, cross-platform
- tags: whisper, transcription, subtitles, forced-alignment, timestamps, srt, asr, natural-language-processing, audio

## Member repositories
- jianfch/stable-ts (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:33.826223+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:41:21.588643+00:00, confidence not recorded.
  - readme: https://github.com/jianfch/stable-ts (fetched 2026-08-28T04:06:33.826223+00:00, sha ed3cb8c804d1)
  - registry_pypi: https://pypi.org/pypi/stable-ts/json (fetched 2026-08-29T10:21:33.885132+00:00, sha 1516f30f75a4)
- Data as of 2026-08-30T08:39:29.467469+00:00.