import os import torch import argparse import gradio as gr from mailersend import emails from dotenv import load_dotenv from elevenlabs.client import ElevenLabs from elevenlabs import play, save import base64 import psycopg2 from urllib.parse import urlparse, parse_qs # Load environment variables load_dotenv() # Argument parsing parser = argparse.ArgumentParser() parser.add_argument("--share", action='store_true', default=False, help="make link public") args = parser.parse_args() # Initialize ElevenLabs client client = ElevenLabs(api_key=os.environ.get("ELEVENLABS_API_KEY")) device = 'cuda' if torch.cuda.is_available() else 'cpu' output_dir = 'outputs' os.makedirs(output_dir, exist_ok=True) supported_languages = ['zh', 'en'] MAILERSEND_API_KEY = os.getenv("MAILERSEND_API_KEY") MAILERSEND_DOMAIN = os.getenv("MAILERSEND_DOMAIN") MAILERSEND_SENDER_EMAIL = f"noreply@{MAILERSEND_DOMAIN}" MAILERSEND_SENDER_NAME = "Voice Clone App" # List of blocked words BLOCKED_WORDS = ['Kill','hurt','shoot','gun','rifle','AR','semi automatic','knife','blade','sword','punch harm','disrupt','blackmail','steal','bitch','cunt','fuck','freaking','nigger','nigga','niggas','cracker','jew','oriental','fag','faggot','account','money','transfer','urgent','help','scared','policy','frightened','accident','fear','scam','address','social security number','assault','injure','maim','destroy','damage','threaten','intimidate','bully','menace','blackmail','extort','exploit','defame','steal','rob','embezzle','defraud Harass','jerk','idiot','stupid','moron','asshole','con','trick','swindle','defraud','payment','credit card','bank account','urgent','immediate','afraid','phone number','email','password'] # Add more words as needed # Function to check for blocked words def contains_blocked_words(text): return any(word.lower() in text.lower() for word in BLOCKED_WORDS) # Function to send email with downloadable file using MailerSend def send_email_with_file(recipient_email, file_path, subject, body): try: mailer = emails.NewEmail(MAILERSEND_API_KEY) mail_body = {} mail_from = { "name": MAILERSEND_SENDER_NAME, "email": MAILERSEND_SENDER_EMAIL, } recipients = [ { "name": "Recipient", "email": recipient_email, } ] mailer.set_mail_from(mail_from, mail_body) mailer.set_mail_to(recipients, mail_body) mailer.set_subject(subject, mail_body) mailer.set_html_content(f"
{body}
", mail_body) mailer.set_plaintext_content(body, mail_body) with open(file_path, "rb") as file: attachment_content = base64.b64encode(file.read()).decode('utf-8') attachments = [ { "filename": os.path.basename(file_path), "content": attachment_content, "disposition": "attachment" } ] mailer.set_attachments(attachments, mail_body) response = mailer.send(mail_body) if response[0] == 202: print("Email sent successfully") return True else: print(f"Failed to send email. Status code: {response[0]}") print(f"Response: {response[1]}") return False except Exception as e: print(f"An error occurred while sending email: {e}") return False # Database connection details connection_string = os.environ.get("DATABASE_URL") result = urlparse(connection_string) user = result.username password = result.password host = result.hostname port = result.port database = result.path[1:] sslmode = parse_qs(result.query)['sslmode'][0] # Function to add user information to the database def add_user_info_to_db(email, order_name, sample_audio, name): connection = psycopg2.connect( dbname=database, user=user, password=password, host=host, port=port, sslmode=sslmode ) cursor = connection.cursor() insert_query = """ INSERT INTO example_table (email, order_name, sample_audio, name) VALUES (%s, %s, %s, %s); """ try: cursor.execute(insert_query, (email, order_name, sample_audio, name)) connection.commit() print("User information added to the database successfully") except Exception as error: print(f"Error occurred during data insertion: {error}") connection.rollback() finally: cursor.close() connection.close() # Predict function def predict(prompt, style, audio_file_pth, voice_name, customer_email, order_name): text_hint = 'Your file will only be saved for 24 hours.\n' if len(prompt) < 2: text_hint += "[ERROR] Please provide a longer prompt text.\n" return text_hint, None, None if len(prompt) > 200: text_hint += "[ERROR] Text length limited to 200 characters. Please try shorter text.\n" return text_hint, None, None if contains_blocked_words(prompt): text_hint += "[ERROR] Your text contains blocked words. Please remove them and try again.\n" return text_hint, None, None full_voice_name = f"{voice_name} - {customer_email}" print(audio_file_pth) voice = client.clone( name=full_voice_name, description="A trial voice model for testing", files=[audio_file_pth], ) audio = client.generate(text=prompt, voice=voice) save_path = f'{output_dir}/output.wav' save(audio, save_path) subject = "Your Voice Clone File" body = "Thank you for using our Voice Clone service. Your file is attached." if send_email_with_file(customer_email, save_path, subject, body): text_hint += "Email sent successfully with the voice file.\n" else: text_hint += "Failed to send email with the voice file. Please try again later.\n" # Add user information to the database with open(audio_file_pth, "rb") as f: sample_audio = f.read() add_user_info_to_db(customer_email, order_name, sample_audio, voice_name) return text_hint, save_path, audio_file_pth # Gradio interface setup with gr.Blocks(gr.themes.Glass()) as demo: with gr.Row(): with gr.Column(): input_text_gr = gr.Textbox( label="Create This", info="One or two sentences at a time is better. Up to 200 text characters.", value="He hoped there would be stew for dinner, turnips and carrots and bruised potatoes and fat mutton pieces to be ladled out in thick, peppered, flour-fattened sauce.", ) style_gr = gr.Dropdown( label="Style", choices=['default', 'whispering', 'cheerful', 'terrified', 'angry', 'sad', 'friendly'], info="Please upload a reference audio file that is at least 1 minute long. For best results, ensure the audio is clear. You can use Adobe Podcast Enhance(https://podcast.adobe.com/enhance) to improve the audio quality before uploading.", max_choices=1, value="default", ) ref_gr = gr.Audio( label="Original Audio", type="filepath", sources=["upload"], ) voice_name_gr = gr.Textbox( label="Your name", value="Sam" ) order_gr = gr.Textbox( label="Your order", value="Sample Order", ) customer_email_gr = gr.Textbox( label="Your Email", info="We'll send you a downloadable file to this email address." ) tts_button = gr.Button("Start", elem_id="send-btn", visible=True) with gr.Column(): out_text_gr = gr.Text(label="Info") audio_gr = gr.Audio(label="Replicated Sound", autoplay=True) ref_audio_gr = gr.Audio(label="Original Audio Used ") tts_button.click(predict, [input_text_gr, style_gr, ref_gr, voice_name_gr, customer_email_gr, order_gr], outputs=[out_text_gr, audio_gr, ref_audio_gr]) demo.queue() demo.launch(debug=True, show_api=False, share=True) css = """ footer {visibility: hidden} audio .btn-container {display: none} """ demo.add_css(css)