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

Whisperlivekit

Whisperlivekit is an open-source Python library and self-hosted server for streaming automatic speech recognition, audio-file transcription, speaker diarization and speech translation. It provides native WebSocket streaming and compatibility-oriented subsets of OpenAI REST and Deepgram WebSocket interfaces.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Whisperlivekit — 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 need a speech-to-text pipeline that runs on infrastructure you operate, especially for live audio, file transcription, diarization, translation or application integration.

Who it is for

  • Developers adding live transcription to a self-hosted application.
  • Teams generating transcripts with speaker labels from conversations or meetings.
  • Builders creating speech translation interfaces or connecting existing clients through supported API subsets.

Documented uses

  • Convert live audio into incremental speech-to-text results.
  • Transcribe an audio file or generate SRT subtitles.
  • Associate transcript content with speakers through diarization.
  • Request translation to a configured target language.
  • Connect an application through native WebSocket, OpenAI-compatible REST or Deepgram-compatible WebSocket interfaces.
Before you choose

Requirements & limits

The repository is public and licensed under Apache-2.0. The README lists Python 3.11–3.13 and provides pip installation instructions.

  • It is a library and server to run yourself, not a hosted transcription service.
  • The documented API interfaces are compatibility-oriented subsets; check the API reference for supported options and current SDK setup.
  • CPU PyTorch and CUDA 12.9 stacks are listed as optional installation choices; other optional backends include Apple Silicon MLX, FunASR and Voxtral MLX.
  • Translation requires an optional dependency, and a translation target must be configured on the server before a session can request one.
  • WebSocket diff mode is explicitly experimental; the bundled web UI uses full-state mode.

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

Install and start a local transcription server

Set up the package and launch its documented local server interface.

Before you start

  • Python 3.11–3.13, as listed by the project
  • Install Whisperlivekit with pip

Steps

  1. Install the package with the command below.

    Documented command
    pip install whisperlivekit

  2. Start the server with wlk --model base --language en.

  3. Open http://localhost:8000 and start talking.

Expected outcome

A local server interface for live speech transcription.

Guide limitations

  • The selected model and language are examples; runtime requirements depend on the backend and model.
  • This recipe covers the documented local startup path, not production deployment or performance guarantees.
Documentation 1 ↗
Source-based practical guide

Transcribe a file and create an SRT subtitle file

Use the command-line interface to transcribe an audio file or write its output as subtitles.

Before you start

  • Whisperlivekit installed, with its command-line tool available
  • An audio file such as WAV or MP3

Steps

  1. Run wlk transcribe meeting.wav to transcribe a file without starting a server.

  2. For subtitles, run wlk transcribe --format srt podcast.mp3 -o podcast.srt.

Expected outcome

A file transcription, or an SRT subtitle file at the specified output path.

Guide limitations

  • The examples document file transcription and SRT output; they do not specify model accuracy or processing time.
Documentation 1 ↗
Evidence and transparency

Sources & review

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