A newer version of the Gradio SDK is available:
5.13.2
metadata
license: apache-2.0
title: Speech 2 Text Kazakh
sdk: gradio
emoji: 🚀
colorFrom: yellow
colorTo: yellow
sdk_version: 5.12.0
This is a Gradio web application that uses Meta's SeamlessM4T model to transcribe speech to text. The app supports both file uploads and microphone input.
Features
- Real-time speech-to-text transcription
- Support for microphone input and file uploads
- GPU acceleration when available
- Simple and intuitive interface
Usage
- Open the app in your browser
- Either upload an audio file or record using your microphone
- Wait for the transcription to appear
Technical Details
- Model: facebook/seamless-m4t-v2-large
- Framework: Gradio
- Backend: PyTorch, Transformers