Home/LLM Frameworks/WhisperJAV
WhisperJAV logo

WhisperJAV

Enrichment pending
meizhong986/WhisperJAV

ASR/STT subtitle generator. Uses Qwen3-ASR, local LLM, Whisper, TEN-VAD. Noise-robust for JAV

GraphCanon updated today · GitHub synced today

1.8k
Stars
159
Forks
122
Open issues
9
Watchers
2mo
Last push
Python MITCreated Oct 18, 2023

Trust & integrity

Full report
Maintenance
Steady (61d since push)
As of today · Source: github_public_v1
Provenance
Not a fork · Personal account
As of today · Source: github_public_v1
Security (OSV)
No lockfile
As of today · Source: none

Public GitHub metadata and optional OSV dependency scans. Signals, not a guarantee. Trust methodology.

Overview

ASR/STT subtitle generator. Uses Qwen3-ASR, local LLM, Whisper, TEN-VAD. Noise-robust for JAV

Capability facts

CLI
CLI entrypoint

Source: pyproject.toml:[project.scripts] · Jul 11, 2026

Languages
python

Source: github.language+pyproject.toml · Jul 11, 2026

Categories

Compatibility

Sourced claims from the README excerpt - not unsourced marketing copy.

Python runtimePython

Source: README excerpt (regex_v1, Jul 11, 2026)

For people who manage their own Python environments.
Source link

Tags

README

Installation

Upgrading? Run whisperjav-upgrade (works on Windows, Linux, and macOS). For code-only updates: whisperjav-upgrade --wheel-only. See the Upgrade Guide for details.



Which Installation Path Should I Follow?

Your SituationGo To
New user on Windows wanting GUIWindows Standalone Installer
Developer on WindowsWindows Source Install
macOS usermacOS (Apple Silicon)
Linux userLinux
Colab or KaggleCloud Notebooks
Developer or expert wanting pipExpert Installation


Windows Source Install

For people who manage their own Python environments.

Prerequisites: Python 3.10-3.12, Git, FFmpeg in PATH.

git clone https://github.com/meizhong986/whisperjav.git
cd whisperjav

:: Full automated install (auto-detects GPU)
installer\install_windows.bat

:: Or with options:
installer\install_windows.bat --cpu-only        :: Force CPU
installer\install_windows.bat --cuda118         :: Force CUDA 11.8
installer\install_windows.bat --cuda128         :: Force CUDA 12.8
installer\install_windows.bat --local-llm       :: Include local LLM translation

The installer runs in 5 phases: PyTorch first (with GPU detection), then scientific stack, Whisper packages, audio/CLI tools, and optional extras. This order matters — PyTorch must be installed before anything that depends on it, or you end up with CPU-only wheels.

For the full walkthrough, see docs/en/guides/installation_windows_python.md.



Recommended: use the install script

chmod +x installer/install_linux.sh ./installer/install_linux.sh


Expert Installation

For users comfortable with Python package management. Choose the components you need.

Modular Installation

WhisperJAV supports modular extras. Install only what you need:


---

# Install in editable mode with dev dependencies
pip install -e ".[dev]"

---

### System Requirements

| Requirement | Minimum | Recommended |
|-------------|---------|-------------|
| **OS** | Windows 10, macOS 11, Ubuntu 20.04 | Windows 11, macOS 14, Ubuntu 22.04 |
| **Python** | 3.10 | 3.11 |
| **RAM** | 8 GB | 16 GB |
| **Disk** | 8 GB | 15 GB (with models) |
| **GPU** | None (CPU works) | NVIDIA RTX 2060+ or Apple Silicon |

**GPU Support:**
- NVIDIA: CUDA 11.8 or 12.8 (Windows, Linux)
- Apple Silicon: MPS acceleration for Whisper (M1/M2/M3/M4/M5). Qwen pipeline is CPU-only on Mac for now.
- AMD ROCm: Experimental (Linux only)
- CPU fallback: Works on all platforms, 5-10x slower

**Processing Time Estimates (per hour of video):**

| Hardware | Time |
|----------|------|
| NVIDIA RTX GPU | 5-10 minutes |
| Apple Silicon | 8-15 minutes |
| CPU | 30-60 minutes |

---

---

## License

MIT License. See [LICENSE](LICENSE) file.

---