https://github.com/noxs1d/speech-to-text

🤖ML project which record audio and converts it to text
https://github.com/noxs1d/speech-to-text

ai fastapi ml torch whisper

Last synced: 3 months ago
JSON representation

🤖ML project which record audio and converts it to text

• Host: GitHub
• URL: https://github.com/noxs1d/speech-to-text
• Owner: noxs1d
• Created: 2024-11-30T21:04:21.000Z (over 1 year ago)
• Default Branch: Release1.0
• Last Pushed: 2025-02-05T18:56:18.000Z (over 1 year ago)
• Last Synced: 2025-04-01T04:38:50.494Z (about 1 year ago)
• Topics: ai, fastapi, ml, torch, whisper
• Language: Jupyter Notebook
• Homepage:
• Size: 5.91 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Speech-to-Text System

___

This project is a Speech-to-Text application built with Python, designed to process audio files, perform transcription, and support functionalities such as recording, uploading, playing audio, and converting it to text.

![architecture](SPEECH-TO-TEXT.png)

## Features

1. Record Audio: Users can record audio directly through the interface. 

2. Upload Audio File: Users can upload an existing audio file for processing.

3. Play Uploaded Audio: The system allows users to listen to uploaded audio files.

4. Convert Audio to Text: Audio files are transcribed into text using state-of-the-art machine learning models.

## Technologies Used

This project leverages the following frameworks and libraries:

- FastAPI: For building a fast, modern, and asynchronous web API.

- Whisper: OpenAI's speech recognition model for transcription.

- Torch (PyTorch): For loading and running the Whisper model efficiently.

- Wave: For audio file manipulation and playback.