LiveTranslate

LiveTranslate

English | 中文

Real-time audio translation for Windows. Captures system audio (WASAPI loopback) and optional microphone input, runs ASR, translates via LLM API, and displays results in a transparent overlay.

Works with any system audio — videos, livestreams, voice chat. No player modifications needed.

Screenshot

Video

Features

• Real-time pipeline: System audio → VAD → ASR → LLM translation → overlay
• Multiple ASR engines: faster-whisper, SenseVoice, FunASR Nano, Anime-Whisper
• Remote ASR: offload speech recognition to a GPU machine over HTTP — see REMOTE_ASR.md
• Any OpenAI-compatible API: DeepSeek, Grok, Qwen, GPT, Ollama, vLLM, etc.
• Streaming translation display: Real-time character-by-character translation output
• Per-model settings: Streaming, structured output (JSON), context history, disable thinking
• Microphone mix-in: Optionally mix microphone input with system audio for ASR
• Low-latency VAD: 32ms chunks + Silero VAD with adaptive silence detection
• Transparent overlay: Always-on-top, click-through, draggable, 14 color themes
• CUDA acceleration: GPU-accelerated ASR inference
• Auto model management: Setup wizard, ModelScope / HuggingFace dual sources
• Built-in benchmark: Compare translation model speed and quality

Changelog

See English Changelog | 中文更新日志

Requirements

• OS: Windows 10/11
• Python: 3.10–3.12 (or use the portable build)
• GPU (recommended): NVIDIA + CUDA 12.6 (Blackwell GPUs like RTX 50xx require CUDA 12.8)
• Network: Access to a translation API

Quick Start

Portable build (no Python required, recommended for non-developers)

Download LiveTranslate-portable-*.zip from Releases, unzip, and double-click start.bat. The first run auto-downloads a portable Python 3.12 and installs GPU-aware dependencies — no Python installation needed.

From source

git clone https://github.com/TheDeathDragon/LiveTranslate.git
cd LiveTranslate

Double-click install.bat — the installer will:

1. Detect Python 3.10–3.12 (auto-install via winget if missing)
2. Create a virtual environment
3. Auto-detect NVIDIA GPU and let you choose CUDA / CPU PyTorch
4. Install all dependencies

Then double-click start.bat to launch.

To update, double-click update.bat — it will pull the latest code and update dependencies (auto-installs Git via winget if missing).

python -m venv .venv
.venv\Scripts\activate

# PyTorch (choose one)
pip install torch torchaudio --index-url https://download.pytorch.org/whl/cu126  # CUDA
pip install torch torchaudio --index-url https://download.pytorch.org/whl/cu128  # CUDA (RTX 50xx)
pip install torch torchaudio --index-url https://download.pytorch.org/whl/cpu    # CPU only

# Dependencies
pip install -r requirements.txt

# Launch
.venv\Scripts\python.exe main.py

First Launch

1. Setup wizard appears — choose download source (ModelScope / HuggingFace) and cache path
2. Silero VAD + SenseVoice models download automatically (~1GB)
3. Main UI appears when ready

Translation API

Settings → Translation tab:

Parameter	Example
API Base	https://api.deepseek.com/v1
API Key	Your key
Model	deepseek-chat
Proxy	none / system / custom URL

Architecture

Audio (WASAPI 32ms) → VAD (Silero) → ASR → LLM Translation → Overlay
         ↑ optional mic mix-in

main.py                 Entry point & pipeline
├── audio_capture.py    WASAPI loopback + mic mix-in
├── vad_processor.py    Silero VAD
├── asr_engine.py       faster-whisper backend
├── asr_funasr.py       Unified FunASR model selector backend
├── asr_sensevoice.py   SenseVoice backend
├── asr_funasr_nano.py  FunASR Nano backend
├── asr_anime_whisper.py Anime-Whisper backend (ja anime/galgame)
├── asr_remote.py        Remote Whisper client (→ asr_server.py, see REMOTE_ASR.md)
├── translator.py       OpenAI-compatible client (streaming, JSON schema, context)
├── model_manager.py    Model download & cache
├── subtitle_overlay.py PyQt6 overlay
├── control_panel.py    Settings UI (7 tabs)
├── dialogs.py          Wizard, download & model config dialogs
└── benchmark.py        Translation benchmark

Acknowledgements

• faster-whisper — Whisper inference via CTranslate2
• FunASR — SenseVoice / Fun-ASR-Nano
• Anime-Whisper — Japanese anime/galgame ASR
• Silero VAD — Voice activity detection

Star History

License

MIT License