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

stt-ru

stt-ru is an open-source Russian speech recognition project built around GigaAM-v3 through onnx-asr. It provides a Python library, CLI, web service with a REST API, and Docker images for transcribing recordings.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

stt-ru — 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 stt-ru when you need Russian speech transcription and want to choose between a CLI, Python integration, browser interface or Docker deployment. It is especially relevant when the result needs to become subtitles or a structured batch export.

Who it is for

  • People transcribing Russian calls, lectures or dictation
  • Developers adding Russian speech recognition to Python workflows or applications
  • Teams producing subtitles or structured transcript files from recordings

Documented uses

  • Transcribe one recording or process a directory of recordings
  • Monitor a folder for new recordings to transcribe
  • Retrieve transcript segments with speaker labels and timestamps through the Python library
  • Export transcripts as text, structured data, subtitles or CSV
  • Offer transcription through a web interface or REST API
Before you choose

Requirements & limits

The project is open source under the MIT license. Python package installation requires Python 3.11 or newer; the model downloads from Hugging Face on first run. Docker images include the model, and the project describes offline recognition with those images.

  • The default recognition model is GigaAM-v3, used through onnx-asr; the model documentation also lists other onnx-asr models.
  • CPU operation does not require PyTorch; NVIDIA GPU operation uses ONNX Runtime.
  • Stereo channels are recognized as separate speakers, according to the project description.
  • Output options include TXT, JSON, JSONL, SRT, VTT, ASS and CSV.
  • The ffmpeg extra is documented for M4A, AAC and audio from video when ffmpeg is not installed.

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 stt-ru for command-line transcription

Set up the Python package and transcribe a single audio file to text in the console.

Before you start

  • Python 3.11 or newer
  • An audio file to transcribe
  • Internet access on first run to download the approximately 230 MB model

Steps

  1. Install the package with pip install stt-ru.

  2. Run stt-ru transcribe call.wav, replacing the example filename with your file.

Expected outcome

The recognized text is printed in the console.

Guide limitations

  • The documented install path requires Python 3.11 or newer.
  • The model downloads from Hugging Face on first run.
  • The optional ffmpeg extra is listed for M4A, AAC and audio from video when ffmpeg is not installed.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
SergeyShk/Speech-to-Text-Russian · Official homepage title · retained in its source language
How we review sourcesSuggest a correction