Spaces:
Sleeping
Sleeping
Yuvvan Talreja commited on
Commit ·
2b7e789
1
Parent(s): 94bf399
Fixed Config Issue
Browse files
README.md
CHANGED
|
@@ -1,52 +1,8 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
## Features
|
| 10 |
-
|
| 11 |
-
- **Audio Transcription**: Powered by OpenAI's Whisper model, providing accurate transcriptions with timestamps
|
| 12 |
-
- **Topic Segmentation**: Automatically identifies topic boundaries and creates hierarchical topic structures
|
| 13 |
-
- **Interactive Visualization**: Generates mindmaps to visualize the structure of the content
|
| 14 |
-
- **Video Integration**: Timestamps in the transcript link directly to the corresponding point in the video
|
| 15 |
-
|
| 16 |
-
## How to Use
|
| 17 |
-
|
| 18 |
-
1. **Upload Video**: Upload your video file (supports MP4, WebM, MOV)
|
| 19 |
-
2. **Choose Model**: Select the transcription model (options range from tiny to medium)
|
| 20 |
-
3. **Transcribe**: Start the transcription process
|
| 21 |
-
4. **Analyze**: Generate the topic segmentation and visualization
|
| 22 |
-
5. **Explore**: Navigate through the interactive mindmap and click on nodes to jump to the relevant sections in your video
|
| 23 |
-
|
| 24 |
-
## Models & Technology
|
| 25 |
-
|
| 26 |
-
- **Transcription**: OpenAI's Whisper (available in various sizes for speed/accuracy tradeoffs)
|
| 27 |
-
- **Topic Segmentation**: Custom algorithm combining TF-IDF, cosine similarity, and hierarchical clustering
|
| 28 |
-
- **Frontend**: HTML/CSS/JS with a responsive design
|
| 29 |
-
- **Backend**: FastAPI for efficient API endpoints
|
| 30 |
-
|
| 31 |
-
## API Endpoints
|
| 32 |
-
|
| 33 |
-
- `POST /transcribe`: Transcribe audio from base64-encoded data
|
| 34 |
-
- `POST /analyze`: Analyze transcript to extract topics
|
| 35 |
-
- `POST /check_model`: Check if a model is available
|
| 36 |
-
|
| 37 |
-
## Running Locally
|
| 38 |
-
|
| 39 |
-
```bash
|
| 40 |
-
git clone https://huggingface.co/spaces/YOUR_USERNAME/flowify
|
| 41 |
-
cd flowify
|
| 42 |
-
pip install -r requirements.txt
|
| 43 |
-
uvicorn fastapi_app:app --host 0.0.0.0 --port 7860 --reload
|
| 44 |
-
```
|
| 45 |
-
|
| 46 |
-
## License
|
| 47 |
-
|
| 48 |
-
MIT
|
| 49 |
-
|
| 50 |
-
## Credits
|
| 51 |
-
|
| 52 |
-
Built using Hugging Face's transformers library and OpenAI's Whisper model.
|
|
|
|
| 1 |
+
title: Flowify - Audio Transcription & Topic Analysis
|
| 2 |
+
emoji: 🔄
|
| 3 |
+
colorFrom: blue
|
| 4 |
+
colorTo: green
|
| 5 |
+
sdk: docker
|
| 6 |
+
app_port: 7860
|
| 7 |
+
pinned: false
|
| 8 |
+
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|