quantumphysicistsam/SpeechGeneration

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

TTS with Voice Cloning – UI & API

This repository provides a text-to-speech (TTS) application built with Coqui TTS’s voice cloning model. It offers both:

• —A Web UI: Built with Gradio for interactive synthesis.
• —An API: Built with FastAPI for programmatic access.

If no speaker reference file is provided, the application will use your default recorded voice (my_voice.wav) for voice cloning.

Features

• —Voice Cloning TTS: Synthesize speech using Coqui TTS’s multilingual voice cloning model.
• —Multiple Language Support: (Supported languages: en, fr-fr, pt-br)
• —Web UI: Interactively enter text, choose language, and optionally upload a speaker reference.
• —API Endpoint: POST requests to synthesize speech from your applications.
• —Deployment Ready: Easily deployable to Hugging Face Spaces via your GitHub repository.

Files in This Repository

• —app.py Contains the FastAPI application with:
• —An API endpoint at /api/synthesize/
• —A mounted Gradio UI at /ui
• —requirements.txt Lists the required Python packages.
• —my_voice.wav Your recorded voice file (used as the default speaker reference if no file is uploaded).

How It Works

API Endpoint

• —URL: /api/synthesize/
• —Method: POST
• —Form Data Parameters:
• —text (string, required): The text to synthesize.
• —language (string, required): The language code (e.g., en, fr-fr, or pt-br).
• —speaker_file (file, optional): A WAV file of a speaker reference. If not provided, the default my_voice.wav will be used if available.

• —Response: Returns the synthesized speech as a WAV file.

Web UI

• —URL: /ui
• —Use the interactive Gradio interface to enter text, select language, and (optionally) upload a speaker file. Click the button to synthesize and play back the audio.

Running Locally

1. 1.Clone the repository:

   git clone https://github.com/<your-username>/<your-repo-name>.git
   cd <your-repo-name>

2. **Install Dependencies**

pip install -r requirements.txt

1. 1.Run the application

   uvicorn app:app --host 0.0.0.0 --port 8000


4. **Access the UI and API**

UI: Open your browser and navigate to http://localhost:8000/ui API: The API endpoint is available at http://localhost:8000/api/synthesize/

1. 1.Sample Python Client for the API Below is a sample Python script that demonstrates how to call the TTS API endpoint using the "requests" library:

    import requests
    # Replace with your deployed Hugging Face Spaces URL (or localhost for testing)
    API_URL = "http://localhost:8000/api/synthesize/"

    # Prepare the form data
    data = {
        "text": "Hello, this is a test of the TTS API.",
        "language": "en"
    }

    # (Optional) Include a speaker reference file if you want:
    # files = {"speaker_file": open("path/to/your/speaker.wav", "rb")}
    files = {}  # No file provided; the default my_voice.wav will be used

    response = requests.post(API_URL, data=data, files=files)

    if response.status_code == 200:
        # Save the returned WAV file
        with open("output.wav", "wb") as f:
            f.write(response.content)
        print("Synthesized audio saved as output.wav")
    else:
        print("Error:", response.json())