# ahmetoner/whisper-asr-webservice

OpenAI Whisper ASR Webservice API

Repository: https://github.com/ahmetoner/whisper-asr-webservice
Canonical: https://ross.abutalabs.com/products/whisper-asr-webservice
Homepage: https://ahmetoner.github.io/whisper-asr-webservice
Language: Python
License: MIT
License Family: permissive
Topics: automatic-speech-recognition, speech-recognition, speech-to-text, openai-whisper, docker, asr, speech
Last push: 2026-08-09T20:36:38+00:00

## Health v2 (maintenance only)
Score: 93/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 96, release rhythm 84, longevity 100
- inputs: {"age_days": 1441, "days_push": 24, "days_rel": 24, "gap_med": 31.0, "n_releases_24m": 9}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3326, forks 584 (observed 2026-08-28T04:07:56.455532+00:00)

## What it is
A Dockerized REST webservice that wraps OpenAI Whisper (plus Faster Whisper and WhisperX engines) for automatic speech recognition. It exposes a Swagger-documented API that transcribes audio/video files into text, JSON, VTT, SRT, or TSV with optional word-level timestamps, VAD filtering, and speaker diarization.

## Use cases
- self-host a speech-to-text API server
- transcribe audio files to text via REST API
- generate SRT or VTT subtitles from audio
- run Whisper transcription with GPU acceleration in Docker
- add speaker diarization to transcriptions
- convert voice recordings to text for my app
- multilingual speech recognition service

## When to choose
- you want a ready-to-run Docker container exposing Whisper as an HTTP API
- you need multiple output formats like SRT/VTT subtitles or word-level timestamps
- you want to switch between Whisper, Faster Whisper, and WhisperX engines via config
- you need GPU-accelerated transcription on Linux with CUDA

## When to avoid
- you need an embedded or offline CLI transcription tool rather than a web service
- you want to fine-tune or train Whisper models yourself
- you need real-time streaming transcription over websockets
- you run on macOS and require GPU acceleration (Docker GPU passthrough is unsupported there)

## Facets
- artifact type: service
- maturity: active
- function: speech-recognition, http-server, api-framework, machine-learning, llm-inference
- domain: speech-processing, artificial-intelligence, self-hosted, apis
- platform: self-hosted, python, cross-platform
- tags: whisper, asr, speech-to-text, transcription, docker-image, fastapi, subtitles, speaker-diarization, gpu-acceleration, docker, web-server, gpu

## Member repositories
- ahmetoner/whisper-asr-webservice (main) score 93

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:56.455532+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 2026-08-29T18:41:46.373079+00:00, confidence not recorded.
  - readme: https://github.com/ahmetoner/whisper-asr-webservice (fetched 2026-08-28T04:07:56.455532+00:00, sha 8236a9e99248)
  - homepage: https://ahmetoner.github.io/whisper-asr-webservice (fetched 2026-08-29T09:34:52.670559+00:00, sha b357fb96e478)
  - site_page: https://ahmetoner.com/whisper-asr-webservice/changelog (fetched 2026-08-29T09:34:52.682231+00:00, sha d91c2c2ec955)
  - site_page: https://ahmetoner.com/whisper-asr-webservice/run (fetched 2026-08-29T09:34:52.679535+00:00, sha ea266e4365fb)
- Data as of 2026-08-30T08:39:29.467469+00:00.