app.py

import gradio as gr
from outetts.v0_1.interface import InterfaceHF

# Initialize the TTS model interface
interface = InterfaceHF("OuteAI/OuteTTS-0.1-350M")

# Define a function to generate and save TTS output from input text
def generate_tts(text, temperature=0.1, repetition_penalty=1.1, max_length=4096):
    output = interface.generate(
        text=text,
        temperature=temperature,
        repetition_penalty=repetition_penalty,
        max_lenght=max_length
    )
    # Save the output audio to a file
    output.save("output.wav")
    return "output.wav"

# Gradio interface for TTS
demo = gr.Interface(
    fn=generate_tts,
    inputs=[
        gr.Textbox(lines=2, placeholder="Enter text to convert to speech", label="Text"),
        gr.Slider(0.1, 1.0, value=0.1, label="Temperature"),
        gr.Slider(1.0, 2.0, value=1.1, label="Repetition Penalty"),
        gr.Slider(512, 4096, value=4096, step=256, label="Max Length")
    ],
    outputs=gr.Audio(type="file", label="Generated Speech"),
    title="OuteTTS Text-to-Speech Demo",
    description="Convert text to speech using the OuteTTS model."
)

# Launch the Gradio app
demo.launch()