#8080th of 81 in VoiceFor voice, speech and music
stt
Offline speech-to-text web app and API built on faster-whisper
- Stars
- 4.9k
- License
- GPL-3.0
- Last commit
- Jan 2026
- Last release
- Aug 2025
- Language
- Python
4,377 on 20 Apr 20264,850 on 9 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Transcribes audio and video files locally with faster-whisper models (tiny to large-v3) and outputs plain text, JSON or timestamped SRT. It runs a Flask web UI and exposes a POST /api endpoint plus an OpenAI-compatible /v1 transcription endpoint on port 9977. A prebuilt Windows package is provided; Linux and macOS run from source with Python 3.9-3.11.
Who it is for: Self-hosters who want a local Whisper transcription API or simple UI
Strengths
- OpenAI-compatible /v1 transcription endpoint works with the official OpenAI Python client
- Outputs text, JSON or SRT subtitles with timestamps
- Prebuilt Windows package with the tiny model bundled; start.exe opens the UI
- Uses CUDA when configured, with CPU as the default
Weaknesses
- No Docker image or compose file; source install needs venv, ffmpeg and manual model download
- CUDA setup is manual: cuDNN, cuBLAS DLLs and editing set.ini
- Chinese output can come out in Traditional characters
- large-v3 on CPU or under 8 GB VRAM can exhaust memory or crash
What it needs
- GPU optional
- Needs Python 3.9-3.11, ffmpeg, PyTorch, CUDA and cuDNN (for GPU)
- Models: faster-whisper tiny, faster-whisper base, faster-whisper small, faster-whisper medium, faster-whisper large-v3
- port 9977
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Quick answers
Is stt open source?
Yes. Its code is on GitHub under the GPL-3.0 license.
Does stt need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in Voice
See all 81| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing, dictation and audiobooks | 98 out of 100 |
| 2 | handyOffline push-to-talk dictation app that pastes transcripts into any text field | 91 out of 100 |
| 3 | whisper.cppC/C++ speech-to-text engine running OpenAI Whisper models locally | 90 out of 100 |
| #4 | xiaozhi-esp32ESP32 firmware for a voice chatbot with MCP device and cloud tool control | 83 out of 100 |
| #5 | buzzOffline desktop app for transcribing and translating audio with Whisper | 83 out of 100 |