# openai/whisper

Robust Speech Recognition via Large-Scale Weak Supervision

Repository: https://github.com/openai/whisper
Canonical: https://ross.abutalabs.com/products/whisper
Language: Python
License: MIT
License Family: permissive
Last push: 2026-07-28T20:18:29+00:00

## Health v2 (maintenance only)
Score: 68/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 94, release rhythm 16, longevity 100
- inputs: {"age_days": 1447, "days_push": 36, "days_rel": 434, "gap_med": 135.0, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 107981, forks 13102 (observed 2026-08-28T04:12:23.793538+00:00)

## What it is
OpenAI's Whisper is a general-purpose speech recognition model and Python library built on a Transformer sequence-to-sequence architecture. It performs multilingual speech transcription, speech translation, and language identification from diverse audio inputs.

## Use cases
- transcribe audio recordings to text
- add subtitles to videos automatically
- translate foreign-language speech to English
- detect the spoken language in an audio clip
- transcribe podcasts or meeting recordings
- run offline speech-to-text without cloud APIs

## When to choose
- you need robust, multilingual speech-to-text with a permissive MIT license
- you want local/offline transcription without sending audio to a cloud service
- you need speech translation or language identification alongside transcription

## When to avoid
- you need real-time low-latency streaming transcription out of the box
- you lack the compute (ideally GPU) for larger model sizes
- you need speaker diarization, which Whisper does not provide

## Facets
- artifact type: library
- maturity: stable
- function: speech-recognition, machine-learning, deep-learning, nlp, cli
- domain: speech-processing, machine-learning, artificial-intelligence
- platform: python, cross-platform, cli
- tags: asr, transcription, speech-translation, language-identification, transformer, pytorch, multilingual, natural-language-processing, gpu

## Member repositories
- openai/whisper (main) score 68

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:23.793538+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 2026-08-29T16:12:10.022286+00:00, confidence not recorded.
  - readme: https://github.com/openai/whisper (fetched 2026-08-28T04:12:23.793538+00:00, sha f31e859d3e9b)
- Data as of 2026-08-30T08:39:29.467469+00:00.