KittenTTS Example

This project demonstrates how to use the KittenTTS text-to-speech model to convert text into a .wav audio file.

Features

• Convert any given text into natural-sounding speech.
• Choose a specific voice for the generated audio.
• Control the playback speed.

Requirements

Install the required Python packages:

apt install espeak
pip install KittenTTS

Usage

Below is a minimal working example:

import soundfile as sf
from kittentts import KittenTTS  
# Initialize the model
model = KittenTTS()

# Generate speech
audio_data = model.generate(
    text="hi who are you",
    voice="expr-voice-2-m",
    speed=1.0
)

# Save output as a WAV file
sf.write("test.wav", audio_data, 24000)

Parameters

• text (str): The text you want to convert to speech.
• voice (str): The voice preset to use (e.g., "expr-voice-2-m").
• speed (float): Playback speed multiplier (1.0 = normal speed).

Reference

https://github.com/KittenML/KittenTTS

Implementation with GUI

https://huggingface.co/spaces/shethjenil/KittenTTS