#3838th of 62 in VoiceFor image, video and design, video and animation, voice, speech and music
funclip
Clip videos by selecting text or speakers from an ASR transcript
- Stars
- 6.4k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Sep 2026
- Language
- Python
5,428 on 2 Apr 20266,448 on 22 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
FunClip runs FunASR Paraformer speech recognition on a video, then lets you pick transcript segments or speaker IDs and cuts the matching clips, with SRT subtitles for the full video and the clip. It runs as a local Gradio app on port 7860, with a command-line mode and optional LLM-assisted segment selection. Backends include Paraformer, Fun-ASR-Nano, SenseVoice and a MOSS vLLM path.
Who it is for: Video editors and developers clipping spoken video by transcript or speaker
Strengths
- Paraformer gives token-level timestamps, so clips cut precisely on selected text
- CAM++ speaker recognition allows clipping all segments from one speaker
- Gradio UI plus a two-stage CLI (recognize, then clip) for scripting
- Optional LLM clipping via OpenAI-compatible gateways or TwelveLabs Pegasus video model
Weaknesses
- Default models are tuned for Chinese; English uses a separate Paraformer model via -l en
- MOSS backend has only segment-level timestamps and needs a separate vLLM service
- No Docker setup in the repo; Python environment and PyTorch must be installed manually
- Reverse period selection and silence removal are still unchecked on the roadmap
What it needs
- GPU optional
- Needs Python 3.12, PyTorch and torchaudio, funasr>=1.4.9, Gradio 4, Pillow, vLLM (MOSS backend only)
- Models: Paraformer-Large, SeACo-Paraformer, CAM++, Fun-ASR-Nano, SenseVoice
- port 7860
Quick answers
Is funclip open source?
Yes. Its code is on GitHub under the MIT license.
Does funclip need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in Voice
See all 62| Rank | Project | Score |
|---|---|---|
| 1 | voicestudioDesktop app for voice cloning, dubbing, dictation and audiobooks | 96 out of 100 |
| 2 | whisper.cppC/C++ port of OpenAI Whisper for local speech-to-text transcription | 89 out of 100 |
| 3 | xiaozhi-esp32ESP32 firmware for a voice chatbot with MCP device and cloud tool control | 82 out of 100 |
| #4 | buzzOffline desktop app for transcribing and translating audio with Whisper | 80 out of 100 |
| #5 | agentsPython framework for realtime voice and multimodal agents on LiveKit WebRTC | 79 out of 100 |