Skip to content

Latest commit

 

History

History
110 lines (77 loc) · 2.58 KB

README.md

File metadata and controls

110 lines (77 loc) · 2.58 KB

Speech-to-Text Transcription App

This project is a web-based application that transcribes audio files into text using Azure Speech-to-Text SDK. It is built with Python, powered by Streamlit, and deployed on Heroku.

Features

  • Upload audio files for transcription.
  • Utilizes Azure Speech-to-Text SDK for accurate and fast transcriptions.
  • Simple and user-friendly web interface created with Streamlit.
  • Secure and scalable deployment on Heroku.

App Screenshot

Here's how the app looks:

App Screenshot

Getting Started

Prerequisites

To run this project locally, ensure you have the following installed:

Installation

  1. Clone the repository:

    git clone https://github.com/s-shahpouri/your-repo-name.git
cd your-repo-name

  2. Create and activate a virtual environment:

    python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

  3. Install the required packages:

    pip install -r requirements.txt

  4. Set up your .env file:

    • Create a .env file in the root directory.
    • Add your Azure Speech-to-Text key and region: 
      AZURE_SPEECH_KEY=your_azure_key

Local Usage

  1. Run the Streamlit app:

    streamlit run app.py

  2. Open the provided URL in your browser (e.g., http://localhost:8501).

  3. Upload an audio file and click the "Transcribe" button to see the transcription.

Deployment

  1. Install the Heroku CLI and log in:

    heroku login

  2. Create a Heroku app:

    heroku create your-app-name

  3. Set your environment variables on Heroku:

    heroku config:set AZURE_SPEECH_KEY=your_azure_key AZURE_REGION=your_azure_region

  4. Push your code to Heroku:

    git push heroku main

Built With

Contributing

Contributions are welcome! Please fork the repository and create a pull request with your changes.

License

This project is licensed under the MIT License. See the LICENSE file for details.