Skip to content
All apps

Whisper speech-to-text

Speech-to-text API for audio and video files using Whisper

Start this app
SourceHomepage
Setup check
Not yet
Setup
One click
Memory
From 2 GB
Category
Model tools
License
MIT

About this app

Whisper ASR Webservice transcribes and translates audio with OpenAI's Whisper models through a small HTTP API, returning text, SRT or VTT subtitles and JSON with timestamps. It uses faster-whisper so the smaller models run on a CPU.

What this app needs

recipe.json

Settings

  • Whisper modelOptional

After it starts

Open the app for interactive API docs. Send audio to POST /asr as the form field audio_file; add ?output=srt for subtitles.

  • The API has no authentication, so anyone with the URL can use it.
  • The first start downloads the model.
Technical details
Opens
app:9000/docs
Check
GET /docs within 600 s
  • app
    Image
    onerahmet/openai-whisper-asr-webservice:v1.10.0
    Ports
    9000/http
    Data
    cache → /root/.cache
    Internet
    Public internet, private networks blocked
    Limits
    2 CPUs · 2048 MB

Checks by Setup

No current check by Setup; runs on your own server do not change the badge.

Past checks

No checks recorded yet.

For maintainers

Share this app’s recipe and setup button.

Add Setup to your repo

Download setup.jsonView file
Preview of the Badge
[![Runs on Setup](https://usesetup.app/api/badge/whisper-asr)](https://usesetup.app/run?source=https%3A%2F%2Fgithub.com%2Fahmetoner%2Fwhisper-asr-webservice)

Green: Setup passed it within 30 days. Blue: CI reported a pass. Your own runners never change the badge.