OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Audio & Voice AI

Whisper ASR Box

Whisper ASR Box is a self-hosted Python speech-recognition service that exposes Whisper, Faster-Whisper and WhisperX through a REST API. It processes audio and video using FFmpeg integration and can provide transcripts, speech translation, language identification and subtitle-oriented outputs.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Whisper ASR Box — OpenViST project profile in Audio & Voice AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want to run a Whisper-based transcription API on your own Docker or Python infrastructure and choose the engine, model and CPU/GPU device. It also suits workflows that need subtitle formats, word timestamps, voice activity detection or WhisperX diarization.

Who it is for

  • Developers adding speech recognition to an application through a REST API.
  • Teams operating an internal transcription service for audio and video.
  • Builders creating captioning or multilingual transcription workflows.
  • Users who need timestamped or speaker-labeled transcripts from a self-managed service.

Documented uses

  • Transcribe audio or video and return text or structured output.
  • Generate VTT or SRT captions from recognized speech.
  • Build speech-translation or language-identification workflows.
  • Produce word-level timestamps and, with WhisperX, speaker diarization.
  • Expose transcription as a REST API for application integration.
Before you choose

Requirements & limits

The repository is public and identifies the software as MIT-licensed. Deployment examples use Docker or a local Python environment.

  • Choose CPU or GPU deployment according to the host setup; the README provides examples for both.
  • Select the ASR engine and model through configuration; supported engines include OpenAI Whisper, Faster-Whisper and WhisperX.
  • Speaker diarization is specifically associated with WhisperX.
  • A persistent cache mount is documented as a way to avoid repeated model downloads and reduce container startup time.
  • Swagger documentation is available at the service's local address after startup.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Start the API in a CPU Docker container

Launch the documented service container with a selected Whisper model and engine.

Before you start

  • Docker
  • A host capable of running the container

Steps

  1. Run the repository's CPU Docker command, mapping port 9000 and setting ASR_MODEL=base and ASR_ENGINE=openai_whisper.

  2. Open the service at localhost:9000 to access its Swagger UI and inspect the API documentation.

Expected outcome

A CPU-configured service container is launched with the Swagger UI available at the documented local address.

Guide limitations

  • This setup recipe does not document a transcription request or an output result.
  • The example uses the base model and OpenAI Whisper engine; other engine and model choices are configuration options.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
ahmetoner/whisper-asr-webservice · Official homepage title · retained in its source language
How we review sourcesSuggest a correction