|
--- |
|
title: MyIVR |
|
emoji: 🎧 |
|
colorFrom: indigo |
|
colorTo: blue |
|
sdk: gradio |
|
sdk_version: 5.12.0 |
|
app_file: app.py |
|
pinned: false |
|
--- |
|
|
|
# MyIVR: Audio Transcription and Summarization |
|
|
|
This Space provides a simple web-based interface to upload audio files, transcribe them into text, and generate a summarized version of the transcription. |
|
|
|
## How It Works |
|
1. Upload an audio file in formats like `.wav` or `.mp3`. |
|
2. The app uses: |
|
- [Whisper](https://github.com/openai/whisper) for audio transcription. |
|
- [Hugging Face Transformers](https://huggingface.co/docs/transformers/index) for text summarization. |
|
3. Get the transcription and summary displayed on the interface. |
|
|
|
## Features |
|
- **Audio Transcription**: Convert speech to text. |
|
- **Text Summarization**: Generate concise summaries of the transcription. |
|
|
|
## Models Used |
|
- **Whisper** (OpenAI): Handles the audio-to-text conversion. |
|
- **T5-Small** (Hugging Face): Summarizes the transcription. |
|
|
|
## Requirements |
|
Dependencies are listed in the `requirements.txt` file. The app uses: |
|
- `gradio` |
|
- `openai-whisper` |
|
- `transformers` |
|
- `torch` |
|
|
|
## Usage |
|
Visit the deployed app at the following link: |
|
[MyIVR on Hugging Face Spaces](https://huggingface.co/spaces/JabriA/MyIVR) |
|
|
|
Upload an audio file, and the app will process it to provide a transcription and summary. |
