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

Voicestreamai

Voicestreamai is an open-source Python server and JavaScript client for streaming audio over WebSockets and transcribing speech near real time with voice activity detection and Whisper-based recognition.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Voicestreamai — 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 self-hosted, WebSocket-based starting point for near-real-time transcription from a JavaScript browser client.

Who it is for

  • Developers building browser-based audio applications that need streaming speech-to-text.
  • Teams adapting voice activity detection, recognition models or audio-chunking strategies in an audio-processing workflow.

Documented uses

  • Stream browser audio to a server and receive speech transcription.
  • Choose a VAD option and configure the speech-recognition model.
  • Set chunk length, offset and transcription language in the included client.
Before you choose

Requirements & limits

The repository identifies the project as MIT licensed. Manual server setup requires Python 3.8 or later; the browser client requires a modern browser with JavaScript support.

  • Faster-whisper is the default ASR implementation; VAD can be set to Silero, Pyannote or none.
  • Silero is documented as token-free, while Pyannote requires an authentication token.
  • Optional certificate and key file arguments support secure WebSocket configuration.
  • The README documents Docker and GPU-oriented use, but says CUDA-in-Docker setup is not covered in detail.

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

Set up the server and connect the browser client

Configure the documented Python server and use the included client to stream audio for transcription.

Before you start

  • Python 3.8 or later and the packages in the project requirements file.
  • A modern web browser with JavaScript support.
  • For Pyannote VAD, a token from the pyannote/segmentation model page; the Silero VAD option is token-free.

Steps

  1. Install the server packages with the command below.

    Documented command
    pip install -r requirements.txt

  2. Start the server with the desired VAD and ASR command-line options; the README gives Silero and Pyannote examples.

  3. Open client/index.html in a browser and enter the server's WebSocket address.

  4. Set audio chunk length and offset, select a transcription language, connect, and use Start Streaming and Stop Streaming to control capture.

Expected outcome

The browser client streams audio to the configured server for speech recognition and transcription.

Guide limitations

  • The guide documents setup and client controls; it does not establish performance for a particular deployment.
  • Pyannote VAD requires an authentication token.
  • The README's Docker example does not explain CUDA-in-Docker setup in detail.
Documentation 1 ↗
Evidence and transparency

Sources & review

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