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

Speech To Text

Speech To Text is a locally run desktop application that converts microphone audio and supported audio files into text using Faster-Whisper, with Silero VAD to detect speech segments. Its HTML-based interface provides transcription and model settings, while additional options include SRT export, WebSocket delivery of results and optional OpenAI API proofreading.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Speech To Text — 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 a locally run Faster-Whisper workflow with configurable model and voice-activity settings, or need to export subtitle files or pass live transcription results to a WebSocket client.

Who it is for

  • People who want to transcribe microphone audio in a desktop application.
  • Users who need text transcripts or SRT subtitles from audio files.
  • Live-stream producers seeking to forward transcription results to a WebSocket client for subtitles.

Documented uses

  • Transcribe speech captured from a microphone.
  • Transcribe supported audio files and export results as SRT subtitles.
  • Forward transcription results from a WebSocket server to a client.
  • Review transcripts alongside synchronized audio and word highlighting when word timestamps are enabled.
  • Optionally proofread transcription text using the OpenAI API.
Before you choose

Requirements & limits

The repository is public and identifies its license as MIT. It documents installation with pip and a Windows run.bat setup.

  • It uses Sounddevice for microphone input and Silero VAD to identify speech segments.
  • Audio-file format support depends partly on Soundfile; WAV is documented, with MP3 and OGG also listed subject to that dependency.
  • A selected local model is referenced from the local folder when a model with the same name is present.
  • OpenAI API proofreading is optional and requires an environment variable containing the API key.
  • The reported sub-second result applies to well-separated sentences on the specified CUDA 11.7 and RTX 3060 12 GB setup.

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 configure Speech To Text

Set up the application and prepare its transcription settings.

Before you start

  • Python package installation with pip; on Windows, use the documented run.bat setup instead.

Steps

  1. Install with pip install ., or on Windows run run.bat to create a virtual environment, install packages and start the application.

  2. Launch with python -m speech_to_text if using the command-line launch method.

  3. In the interface, configure App Settings, Model Settings, Transcribe Settings and VAD Settings.

  4. Start Transcription.

Expected outcome

The application is configured and transcription can be started from its interface.

Guide limitations

  • The guide does not specify hardware requirements or guarantee performance across machines.
  • OpenAI-based proofreading is optional and requires OPENAI_API_KEY as an environment variable.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
reriiasu/speech-to-text · Official homepage title · retained in its source language
How we review sourcesSuggest a correction