app.py

import gradio as gr
import requests
import json
import os

# Replace with your actual Play.ht API key and User ID
API_KEY = os.getenv('PLAY_API_KEY')
USER_ID = os.getenv('PLAY_USER_ID')

# Function to interact with Play.ht API
def text_to_audio(text):
    url = "https://play.ht/api/v2/tts"

    headers = {
        "Authorization": API_KEY,
        "X-User-ID": USER_ID,
        "Content-Type": "application/json"
    }

    # Customize options here (e.g., voice, language, etc.)
    data = {
        "voice": "en_us-male-1",  # Specify a voice of choice
        "content": [text],
        "speed": 1.0,
        "sample_rate": 24000,
    }

    response = requests.post(url, headers=headers, json=data)
    
    if response.status_code == 201:
        response_data = response.json()
        audio_url = response_data.get("audio_url")
        
        # Fetch the audio file
        audio_response = requests.get(audio_url)
        audio_path = "output_audio.mp3"
        with open(audio_path, "wb") as audio_file:
            audio_file.write(audio_response.content)
        
        return audio_path
    else:
        return f"Error: {response.status_code}, {response.text}"

# Gradio app interface
iface = gr.Interface(
    fn=text_to_audio,
    inputs="text",
    outputs="audio",
    title="Text to Speech with Play.ht",
    description="Convert text into realistic speech using Play.ht."
)

iface.launch(debug = True)