audio.py

import random 
import numpy as np 
from elevenlabs import voices, generate, set_api_key, UnauthenticatedRateLimitError, save
import huggingface_hub
from huggingface_hub import Repository
import os

DATASET_REPO_URL = "https://huggingface.co/datasets/laxsvips/audiofiles"
DATA_FILENAME = "audio.mp4"
DATA_FILE = os.path.join("data", DATA_FILENAME)

# HF_TOKEN = os.environ.get("HF_TOKEN")
# repo = Repository(
#     local_dir="data", clone_from=DATASET_REPO_URL, use_auth_token=HF_TOKEN
# )

def pad_buffer(audio):
    # Pad buffer to multiple of 2 bytes
    buffer_size = len(audio)
    element_size = np.dtype(np.int16).itemsize
    if buffer_size % element_size != 0:
        audio = audio + b'\0' * (element_size - (buffer_size % element_size))
    return audio 

def generate_voice(text):
    try:
        audio = generate(
            text, 
            voice="Arnold", 
            model="eleven_monolingual_v1"
        )
        save(audio,'data/audio.mp4')        
        return (44100, np.frombuffer(pad_buffer(audio), dtype=np.int16))
    except UnauthenticatedRateLimitError as e:
        raise gr.Error("Thanks for trying out ElevenLabs TTS! You've reached the free tier limit. Please provide an API key to continue.") 
    except Exception as e:
        raise gr.Error(e)