File size: 1,358 Bytes
a457669
 
20292f7
 
 
a457669
 
 
 
 
 
20292f7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
---
title: TextCut
emoji: ✂️
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 6.3.0
app_file: app.py
pinned: false
---

# TextCut

Edit videos by simply editing their transcript. Upload a video, get an automatic transcription with timestamps using VibeVoice-ASR, then delete lines from the transcript to cut those parts from your video.

## Features

- **Automatic Transcription**: Uses Microsoft's VibeVoice-ASR model for accurate speech-to-text with timestamps
- **Real-time Highlighting**: Current sentence is highlighted (uppercased) as the video plays
- **Simple Editing**: Delete lines from the transcript to mark segments for removal
- **Video Cutting**: Automatically cuts the video based on deleted transcript segments using FFmpeg

## Usage

1. **Upload**: Upload a video file (mp4, mov, etc.)
2. **Transcribe**: Click "Transcribe" to generate the transcript with timestamps
3. **Edit**: Delete lines from the transcript that you want to cut from the video
4. **Apply Cuts**: Click "Apply Cuts" to generate the edited video

## Requirements

- Python 3.10+
- FFmpeg installed on the system
- CUDA-capable GPU (for transcription)

## Local Development

```bash
pip install -r requirements.txt
python app.py
```

## Hugging Face Spaces

This app is designed to run on Hugging Face Spaces with ZeroGPU support for the transcription model.