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

Transcription Stream Community Edition

Transcription Stream Community Edition is a GPL-3.0-licensed, Python-based self-hosted application for offline audio transcription and speaker diarization, with an Ollama-and-Mistral summary feature. It runs as Docker services and provides a browser interface and SSH drop zones for submitting and reviewing recordings; Meilisearch is included for full-text search.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Transcription Stream Community Edition — 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 can provide an NVIDIA-GPU-equipped Docker host and want an offline, self-managed recording workflow that combines transcripts, speaker labels and LLM-generated summaries.

Who it is for

  • People who want to transcribe and label speakers in recordings on self-managed infrastructure.
  • Teams seeking a browser-based way to upload, review, listen to and download transcription output.
  • Technical users who want generated summaries and full-text search alongside transcripts.

Documented uses

  • Transcribe recordings and identify speakers.
  • Submit audio through a web interface or SSH drop zones.
  • Review time-synced transcript text while listening, then download output.
  • Generate summaries using Ollama and Mistral.
  • Search transcription text with the included Meilisearch service.
Before you choose

Requirements & limits

  • The README specifies an NVIDIA GPU prerequisite and warns the GPU image is about 26GB.
  • The maintainer reports that 12GB of VRAM may not reliably support diarization and Ollama Mistral at the same time; a separate Ollama endpoint is described as a workaround.
  • The project includes Docker scripts for pulling images or building locally.
  • The README cautions that the example code requires additional security measures before production use.
  • The repository identifies the project as GPL-3.0 licensed.

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 Docker services with the no-build script

Pull the Docker images and start the service stack using the documented quick-start path.

Before you start

  • A host with an NVIDIA GPU.
  • The repository files, including start-nobuild.sh.
  • Docker images and sufficient capacity for the approximately 26GB ts-gpu image.

Steps

  1. Run ./start-nobuild.sh to pull the Docker images and start services.

  2. Open the web interface at http://dockerip:5006 on the Docker host.

Expected outcome

The services start and the web interface is available for audio upload, review and download.

Guide limitations

  • The README warns that the ts-gpu image is approximately 26GB.
  • The project README says the example code needs additional security measures before production use.
Documentation 1 ↗
Evidence and transparency

Sources & review

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