|
|
import random |
|
|
import gradio as gr |
|
|
import numpy as np |
|
|
from elevenlabs import voices, generate, set_api_key, UnauthenticatedRateLimitError |
|
|
|
|
|
def pad_buffer(audio): |
|
|
|
|
|
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, voice_name): |
|
|
try: |
|
|
audio = generate( |
|
|
text[:250], |
|
|
voice=voice_name, |
|
|
model="eleven_multilingual_v2" |
|
|
) |
|
|
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) |
|
|
|
|
|
|
|
|
badges = """ |
|
|
|
|
|
""" |
|
|
menu = """ |
|
|
<style> |
|
|
.menu-bar { |
|
|
display: flex; |
|
|
justify-content: space-between; |
|
|
align-items: center; |
|
|
background-color: #333; |
|
|
padding: 10px; |
|
|
color: white; |
|
|
font-family: 'sans-serif'; /* Use sans-serif font */ |
|
|
} |
|
|
.menu-icon { |
|
|
font-size: 24px; |
|
|
background-color: #ffffff; |
|
|
border-radius: 50%; |
|
|
padding: 5px; |
|
|
text-decoration: none; /* Remove underline from emoji */ |
|
|
} |
|
|
.menu-items { |
|
|
display: flex; |
|
|
gap: 15px; |
|
|
color: white; |
|
|
} |
|
|
.menu-item { |
|
|
padding: 8px 16px; |
|
|
background-color: #555; |
|
|
border-radius: 4px; |
|
|
transition: background-color 0.3s; |
|
|
font-weight: bold; /* Make text bold */ |
|
|
} |
|
|
.menu-item a, .menu-icon a { |
|
|
color: white; |
|
|
text-decoration: none; /* Remove underline */ |
|
|
} |
|
|
.menu-item:hover { |
|
|
background-color: #777; |
|
|
} |
|
|
</style> |
|
|
|
|
|
<div class="menu-bar"> |
|
|
<a href="#" class="menu-icon">🎵</a> <!-- Make emoji clickable --> |
|
|
<div class="menu-items"> |
|
|
<span class="menu-item"><a href="#">Dashboard</a></span> |
|
|
<span class="menu-item"><a href="#">Premium Apps</a></span> |
|
|
<span class="menu-item"><a href="#">My Account</a></span> |
|
|
<span class="menu-item"><a href="#">Train Voice</a></span> |
|
|
</div> |
|
|
</div> |
|
|
|
|
|
""" |
|
|
|
|
|
description = """ |
|
|
Welcome to AI Tutor Text to Voice, single foundational model supporting 28 languages including: English, Chinese, Spanish, Hindi, Portuguese, French, German, Japanese, Arabic, Korean, Indonesian, Italian, Dutch, Turkish, Polish, Swedish, Filipino, Malay, Romanian, Ukrainian, Greek, Czech, Danish, Finnish, Bulgarian, Croatian, Slovak, and Tamil. Sign up on [ElevenLabs](https://elevenlabs.io) to get fast access, long-form generation, voice cloning, API keys, and more! |
|
|
""" |
|
|
|
|
|
|
|
|
with gr.Blocks() as block: |
|
|
gr.Markdown('') |
|
|
gr.Markdown(menu) |
|
|
gr.Markdown(badges) |
|
|
gr.Markdown(description) |
|
|
|
|
|
input_text = gr.Textbox( |
|
|
label="Input Text (250 characters max)", |
|
|
lines=2, |
|
|
value="Hello! 你好! Hola! नमस्ते! Bonjour! こんにちは! مرحبا! 안녕하세요! Ciao! Cześć! Привіт! Γειά σας! Здравей! வணக்கம்!", |
|
|
elem_id="input_text" |
|
|
) |
|
|
|
|
|
all_voices = voices() |
|
|
input_voice = gr.Dropdown( |
|
|
[ voice.name for voice in all_voices ], |
|
|
value="Bella", |
|
|
label="Voice", |
|
|
elem_id="input_voice" |
|
|
) |
|
|
|
|
|
run_button = gr.Button( |
|
|
text="Generate Voice", |
|
|
type="button" |
|
|
) |
|
|
|
|
|
out_audio = gr.Audio( |
|
|
label="Generated Voice", |
|
|
type="numpy", |
|
|
elem_id="out_audio", |
|
|
format="mp3" |
|
|
) |
|
|
|
|
|
inputs = [input_text, input_voice] |
|
|
outputs = [out_audio] |
|
|
|
|
|
run_button.click( |
|
|
fn=generate_voice, |
|
|
inputs=inputs, |
|
|
outputs=outputs, |
|
|
queue=True |
|
|
) |
|
|
|
|
|
block.queue(concurrency_count=5).launch(debug=True) |
