File size: 34,617 Bytes
518cdf9 6466c67 4a45bb8 6466c67 4a45bb8 518cdf9 d069ecb bc19ae4 6001005 67a6da3 518cdf9 d069ecb 746f439 d069ecb e70790f 6001005 bc19ae4 6001005 518cdf9 4a45bb8 518cdf9 4a45bb8 518cdf9 4a45bb8 518cdf9 d069ecb 518cdf9 d069ecb 518cdf9 d069ecb 518cdf9 4a45bb8 518cdf9 4a45bb8 d069ecb 4a45bb8 d069ecb 4a45bb8 d069ecb 4a45bb8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 |
print("starting...")
import argparse
language_options = [
"en", "es", "fr", "de", "it", "pt", "pl", "tr", "ru", "nl", "cs", "ar", "zh-cn", "ja", "hu", "ko"
]
# Convert the list of languages to a string to display in the help text
language_options_str = ", ".join(language_options)
# Argument parser to handle optional parameters with descriptions
parser = argparse.ArgumentParser(
description="Convert eBooks to Audiobooks using a Text-to-Speech model. You can either launch the Gradio interface or run the script in headless mode for direct conversion.",
epilog="Example: python script.py --headless --ebook path_to_ebook --voice path_to_voice --language en --use_custom_model True --custom_model model.pth --custom_config config.json --custom_vocab vocab.json"
)
parser.add_argument("--share", type=bool, default=False, help="Set to True to enable a public shareable Gradio link. Defaults to False.")
parser.add_argument("--headless", type=bool, default=False, help="Set to True to run in headless mode without the Gradio interface. Defaults to False.")
parser.add_argument("--ebook", type=str, help="Path to the ebook file for conversion. Required in headless mode.")
parser.add_argument("--voice", type=str, help="Path to the target voice file for TTS. Optional, uses a default voice if not provided.")
parser.add_argument("--language", type=str, default="en",
help=f"Language for the audiobook conversion. Options: {language_options_str}. Defaults to English (en).")
parser.add_argument("--use_custom_model", type=bool, default=False,
help="Set to True to use a custom TTS model. Defaults to False. Must be True to use custom models, otherwise you'll get an error.")
parser.add_argument("--custom_model", type=str, help="Path to the custom model file (.pth). Required if using a custom model.")
parser.add_argument("--custom_config", type=str, help="Path to the custom config file (config.json). Required if using a custom model.")
parser.add_argument("--custom_vocab", type=str, help="Path to the custom vocab file (vocab.json). Required if using a custom model.")
parser.add_argument("--custom_model_url", type=str,
help=("URL to download the custom model as a zip file. Optional, but will be used if provided. "
"Examples include David Attenborough's model: "
"'https://huggingface.co/drewThomasson/xtts_David_Attenborough_fine_tune/resolve/main/Finished_model_files.zip?download=true'. "
"More XTTS fine-tunes can be found on my Hugging Face at 'https://huggingface.co/drewThomasson'."))
args = parser.parse_args()
import os
import shutil
import subprocess
import re
from pydub import AudioSegment
import tempfile
from pydub import AudioSegment
import nltk
from nltk.tokenize import sent_tokenize
import sys
import torch
from TTS.api import TTS
from TTS.tts.configs.xtts_config import XttsConfig
from TTS.tts.models.xtts import Xtts
from tqdm import tqdm
import gradio as gr
from gradio import Progress
import urllib.request
import zipfile
import socket
#import MeCab
#import unidic
#nltk.download('punkt_tab')
# Import the locally stored Xtts default model
import import_locally_stored_tts_model_files
#make the nltk folder point to the nltk folder in the app dir
nltk.data.path.append('/home/user/app/nltk_data')
# Download UniDic if it's not already installed
#unidic.download()
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(f"Device selected is: {device}")
#nltk.download('punkt') # Make sure to download the necessary models
def download_and_extract_zip(url, extract_to='.'):
try:
# Ensure the directory exists
os.makedirs(extract_to, exist_ok=True)
zip_path = os.path.join(extract_to, 'model.zip')
# Download with progress bar
with tqdm(unit='B', unit_scale=True, miniters=1, desc="Downloading Model") as t:
def reporthook(blocknum, blocksize, totalsize):
t.total = totalsize
t.update(blocknum * blocksize - t.n)
urllib.request.urlretrieve(url, zip_path, reporthook=reporthook)
print(f"Downloaded zip file to {zip_path}")
# Unzipping with progress bar
with zipfile.ZipFile(zip_path, 'r') as zip_ref:
files = zip_ref.namelist()
with tqdm(total=len(files), unit="file", desc="Extracting Files") as t:
for file in files:
if not file.endswith('/'): # Skip directories
# Extract the file to the temporary directory
extracted_path = zip_ref.extract(file, extract_to)
# Move the file to the base directory
base_file_path = os.path.join(extract_to, os.path.basename(file))
os.rename(extracted_path, base_file_path)
t.update(1)
# Cleanup: Remove the ZIP file and any empty folders
os.remove(zip_path)
for root, dirs, files in os.walk(extract_to, topdown=False):
for name in dirs:
os.rmdir(os.path.join(root, name))
print(f"Extracted files to {extract_to}")
# Check if all required files are present
required_files = ['model.pth', 'config.json', 'vocab.json_']
missing_files = [file for file in required_files if not os.path.exists(os.path.join(extract_to, file))]
if not missing_files:
print("All required files (model.pth, config.json, vocab.json_) found.")
else:
print(f"Missing files: {', '.join(missing_files)}")
except Exception as e:
print(f"Failed to download or extract zip file: {e}")
def is_folder_empty(folder_path):
if os.path.exists(folder_path) and os.path.isdir(folder_path):
# List directory contents
if not os.listdir(folder_path):
return True # The folder is empty
else:
return False # The folder is not empty
else:
print(f"The path {folder_path} is not a valid folder.")
return None # The path is not a valid folder
def remove_folder_with_contents(folder_path):
try:
shutil.rmtree(folder_path)
print(f"Successfully removed {folder_path} and all of its contents.")
except Exception as e:
print(f"Error removing {folder_path}: {e}")
def wipe_folder(folder_path):
# Check if the folder exists
if not os.path.exists(folder_path):
print(f"The folder {folder_path} does not exist.")
return
# Iterate over all the items in the given folder
for item in os.listdir(folder_path):
item_path = os.path.join(folder_path, item)
# If it's a file, remove it and print a message
if os.path.isfile(item_path):
os.remove(item_path)
print(f"Removed file: {item_path}")
# If it's a directory, remove it recursively and print a message
elif os.path.isdir(item_path):
shutil.rmtree(item_path)
print(f"Removed directory and its contents: {item_path}")
print(f"All contents wiped from {folder_path}.")
# Example usage
# folder_to_wipe = 'path_to_your_folder'
# wipe_folder(folder_to_wipe)
def create_m4b_from_chapters(input_dir, ebook_file, output_dir):
# Function to sort chapters based on their numeric order
def sort_key(chapter_file):
numbers = re.findall(r'\d+', chapter_file)
return int(numbers[0]) if numbers else 0
# Extract metadata and cover image from the eBook file
def extract_metadata_and_cover(ebook_path):
try:
cover_path = ebook_path.rsplit('.', 1)[0] + '.jpg'
subprocess.run(['ebook-meta', ebook_path, '--get-cover', cover_path], check=True)
if os.path.exists(cover_path):
return cover_path
except Exception as e:
print(f"Error extracting eBook metadata or cover: {e}")
return None
# Combine WAV files into a single file
def combine_wav_files(chapter_files, output_path, batch_size=256):
# Initialize an empty audio segment
combined_audio = AudioSegment.empty()
# Process the chapter files in batches
for i in range(0, len(chapter_files), batch_size):
batch_files = chapter_files[i:i + batch_size]
batch_audio = AudioSegment.empty() # Initialize an empty AudioSegment for the batch
# Sequentially append each file in the current batch to the batch_audio
for chapter_file in batch_files:
audio_segment = AudioSegment.from_wav(chapter_file)
batch_audio += audio_segment
# Combine the batch audio with the overall combined_audio
combined_audio += batch_audio
# Export the combined audio to the output file path
combined_audio.export(output_path, format='wav')
print(f"Combined audio saved to {output_path}")
# Function to generate metadata for M4B chapters
def generate_ffmpeg_metadata(chapter_files, metadata_file):
with open(metadata_file, 'w') as file:
file.write(';FFMETADATA1\n')
start_time = 0
for index, chapter_file in enumerate(chapter_files):
duration_ms = len(AudioSegment.from_wav(chapter_file))
file.write(f'[CHAPTER]\nTIMEBASE=1/1000\nSTART={start_time}\n')
file.write(f'END={start_time + duration_ms}\ntitle=Chapter {index + 1}\n')
start_time += duration_ms
# Generate the final M4B file using ffmpeg
def create_m4b(combined_wav, metadata_file, cover_image, output_m4b):
# Ensure the output directory exists
os.makedirs(os.path.dirname(output_m4b), exist_ok=True)
ffmpeg_cmd = ['ffmpeg', '-i', combined_wav, '-i', metadata_file]
if cover_image:
ffmpeg_cmd += ['-i', cover_image, '-map', '0:a', '-map', '2:v']
else:
ffmpeg_cmd += ['-map', '0:a']
ffmpeg_cmd += ['-map_metadata', '1', '-c:a', 'aac', '-b:a', '192k']
if cover_image:
ffmpeg_cmd += ['-c:v', 'png', '-disposition:v', 'attached_pic']
ffmpeg_cmd += [output_m4b]
subprocess.run(ffmpeg_cmd, check=True)
# Main logic
chapter_files = sorted([os.path.join(input_dir, f) for f in os.listdir(input_dir) if f.endswith('.wav')], key=sort_key)
temp_dir = tempfile.gettempdir()
temp_combined_wav = os.path.join(temp_dir, 'combined.wav')
metadata_file = os.path.join(temp_dir, 'metadata.txt')
cover_image = extract_metadata_and_cover(ebook_file)
output_m4b = os.path.join(output_dir, os.path.splitext(os.path.basename(ebook_file))[0] + '.m4b')
combine_wav_files(chapter_files, temp_combined_wav)
generate_ffmpeg_metadata(chapter_files, metadata_file)
create_m4b(temp_combined_wav, metadata_file, cover_image, output_m4b)
# Cleanup
if os.path.exists(temp_combined_wav):
os.remove(temp_combined_wav)
if os.path.exists(metadata_file):
os.remove(metadata_file)
if cover_image and os.path.exists(cover_image):
os.remove(cover_image)
# Example usage
# create_m4b_from_chapters('path_to_chapter_wavs', 'path_to_ebook_file', 'path_to_output_dir')
#this code right here isnt the book grabbing thing but its before to refrence in order to create the sepecial chapter labeled book thing with calibre idk some systems cant seem to get it so just in case but the next bit of code after this is the book grabbing code with booknlp
import os
import subprocess
import ebooklib
from ebooklib import epub
from bs4 import BeautifulSoup
import re
import csv
import nltk
# Only run the main script if Value is True
def create_chapter_labeled_book(ebook_file_path):
# Function to ensure the existence of a directory
def ensure_directory(directory_path):
if not os.path.exists(directory_path):
os.makedirs(directory_path)
print(f"Created directory: {directory_path}")
ensure_directory(os.path.join(".", 'Working_files', 'Book'))
def convert_to_epub(input_path, output_path):
# Convert the ebook to EPUB format using Calibre's ebook-convert
try:
subprocess.run(['ebook-convert', input_path, output_path], check=True)
except subprocess.CalledProcessError as e:
print(f"An error occurred while converting the eBook: {e}")
return False
return True
def save_chapters_as_text(epub_path):
# Create the directory if it doesn't exist
directory = os.path.join(".", "Working_files", "temp_ebook")
ensure_directory(directory)
# Open the EPUB file
book = epub.read_epub(epub_path)
previous_chapter_text = ''
previous_filename = ''
chapter_counter = 0
# Iterate through the items in the EPUB file
for item in book.get_items():
if item.get_type() == ebooklib.ITEM_DOCUMENT:
# Use BeautifulSoup to parse HTML content
soup = BeautifulSoup(item.get_content(), 'html.parser')
text = soup.get_text()
# Check if the text is not empty
if text.strip():
if len(text) < 2300 and previous_filename:
# Append text to the previous chapter if it's short
with open(previous_filename, 'a', encoding='utf-8') as file:
file.write('\n' + text)
else:
# Create a new chapter file and increment the counter
previous_filename = os.path.join(directory, f"chapter_{chapter_counter}.txt")
chapter_counter += 1
with open(previous_filename, 'w', encoding='utf-8') as file:
file.write(text)
print(f"Saved chapter: {previous_filename}")
# Example usage
input_ebook = ebook_file_path # Replace with your eBook file path
output_epub = os.path.join(".", "Working_files", "temp.epub")
if os.path.exists(output_epub):
os.remove(output_epub)
print(f"File {output_epub} has been removed.")
else:
print(f"The file {output_epub} does not exist.")
if convert_to_epub(input_ebook, output_epub):
save_chapters_as_text(output_epub)
# Download the necessary NLTK data (if not already present)
#nltk.download('punkt')
def process_chapter_files(folder_path, output_csv):
with open(output_csv, 'w', newline='', encoding='utf-8') as csvfile:
writer = csv.writer(csvfile)
# Write the header row
writer.writerow(['Text', 'Start Location', 'End Location', 'Is Quote', 'Speaker', 'Chapter'])
# Process each chapter file
chapter_files = sorted(os.listdir(folder_path), key=lambda x: int(x.split('_')[1].split('.')[0]))
for filename in chapter_files:
if filename.startswith('chapter_') and filename.endswith('.txt'):
chapter_number = int(filename.split('_')[1].split('.')[0])
file_path = os.path.join(folder_path, filename)
try:
with open(file_path, 'r', encoding='utf-8') as file:
text = file.read()
# Insert "NEWCHAPTERABC" at the beginning of each chapter's text
if text:
text = "NEWCHAPTERABC" + text
sentences = nltk.tokenize.sent_tokenize(text)
for sentence in sentences:
start_location = text.find(sentence)
end_location = start_location + len(sentence)
writer.writerow([sentence, start_location, end_location, 'True', 'Narrator', chapter_number])
except Exception as e:
print(f"Error processing file {filename}: {e}")
# Example usage
folder_path = os.path.join(".", "Working_files", "temp_ebook")
output_csv = os.path.join(".", "Working_files", "Book", "Other_book.csv")
process_chapter_files(folder_path, output_csv)
def sort_key(filename):
"""Extract chapter number for sorting."""
match = re.search(r'chapter_(\d+)\.txt', filename)
return int(match.group(1)) if match else 0
def combine_chapters(input_folder, output_file):
# Create the output folder if it doesn't exist
os.makedirs(os.path.dirname(output_file), exist_ok=True)
# List all txt files and sort them by chapter number
files = [f for f in os.listdir(input_folder) if f.endswith('.txt')]
sorted_files = sorted(files, key=sort_key)
with open(output_file, 'w', encoding='utf-8') as outfile: # Specify UTF-8 encoding here
for i, filename in enumerate(sorted_files):
with open(os.path.join(input_folder, filename), 'r', encoding='utf-8') as infile: # And here
outfile.write(infile.read())
# Add the marker unless it's the last file
if i < len(sorted_files) - 1:
outfile.write("\nNEWCHAPTERABC\n")
# Paths
input_folder = os.path.join(".", 'Working_files', 'temp_ebook')
output_file = os.path.join(".", 'Working_files', 'Book', 'Chapter_Book.txt')
# Combine the chapters
combine_chapters(input_folder, output_file)
ensure_directory(os.path.join(".", "Working_files", "Book"))
#create_chapter_labeled_book()
import os
import subprocess
import sys
import torchaudio
# Check if Calibre's ebook-convert tool is installed
def calibre_installed():
try:
subprocess.run(['ebook-convert', '--version'], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
return True
except FileNotFoundError:
print("Calibre is not installed. Please install Calibre for this functionality.")
return False
import os
import torch
from TTS.api import TTS
from nltk.tokenize import sent_tokenize
from pydub import AudioSegment
# Assuming split_long_sentence and wipe_folder are defined elsewhere in your code
default_target_voice_path = "default_voice.wav" # Ensure this is a valid path
default_language_code = "en"
# Function to check if vocab.json exists and rename it
def rename_vocab_file_if_exists(directory):
vocab_path = os.path.join(directory, 'vocab.json')
new_vocab_path = os.path.join(directory, 'vocab.json_')
# Check if vocab.json exists
if os.path.exists(vocab_path):
# Rename the file
os.rename(vocab_path, new_vocab_path)
print(f"Renamed {vocab_path} to {new_vocab_path}")
return True # Return True if the file was found and renamed
def combine_wav_files(input_directory, output_directory, file_name):
# Ensure that the output directory exists, create it if necessary
os.makedirs(output_directory, exist_ok=True)
# Specify the output file path
output_file_path = os.path.join(output_directory, file_name)
# Initialize an empty audio segment
combined_audio = AudioSegment.empty()
# Get a list of all .wav files in the specified input directory and sort them
input_file_paths = sorted(
[os.path.join(input_directory, f) for f in os.listdir(input_directory) if f.endswith(".wav")],
key=lambda f: int(''.join(filter(str.isdigit, f)))
)
# Sequentially append each file to the combined_audio
for input_file_path in input_file_paths:
audio_segment = AudioSegment.from_wav(input_file_path)
combined_audio += audio_segment
# Export the combined audio to the output file path
combined_audio.export(output_file_path, format='wav')
print(f"Combined audio saved to {output_file_path}")
# Function to split long strings into parts
def split_long_sentence(sentence, max_length=249, max_pauses=10):
"""
Splits a sentence into parts based on length or number of pauses without recursion.
:param sentence: The sentence to split.
:param max_length: Maximum allowed length of a sentence.
:param max_pauses: Maximum allowed number of pauses in a sentence.
:return: A list of sentence parts that meet the criteria.
"""
parts = []
while len(sentence) > max_length or sentence.count(',') + sentence.count(';') + sentence.count('.') > max_pauses:
possible_splits = [i for i, char in enumerate(sentence) if char in ',;.' and i < max_length]
if possible_splits:
# Find the best place to split the sentence, preferring the last possible split to keep parts longer
split_at = possible_splits[-1] + 1
else:
# If no punctuation to split on within max_length, split at max_length
split_at = max_length
# Split the sentence and add the first part to the list
parts.append(sentence[:split_at].strip())
sentence = sentence[split_at:].strip()
# Add the remaining part of the sentence
parts.append(sentence)
return parts
"""
if 'tts' not in locals():
tts = TTS(selected_tts_model, progress_bar=True).to(device)
"""
from tqdm import tqdm
# Convert chapters to audio using XTTS
def convert_chapters_to_audio_custom_model(chapters_dir, output_audio_dir, target_voice_path=None, language=None, custom_model=None):
if target_voice_path==None:
target_voice_path = default_target_voice_path
if custom_model:
print("Loading custom model...")
config = XttsConfig()
config.load_json(custom_model['config'])
model = Xtts.init_from_config(config)
model.load_checkpoint(config, checkpoint_path=custom_model['model'], vocab_path=custom_model['vocab'], use_deepspeed=False)
model.to(device)
print("Computing speaker latents...")
gpt_cond_latent, speaker_embedding = model.get_conditioning_latents(audio_path=[target_voice_path])
else:
selected_tts_model = "tts_models/multilingual/multi-dataset/xtts_v2"
tts = TTS(selected_tts_model, progress_bar=False).to(device)
if not os.path.exists(output_audio_dir):
os.makedirs(output_audio_dir)
for chapter_file in sorted(os.listdir(chapters_dir)):
if chapter_file.endswith('.txt'):
match = re.search(r"chapter_(\d+).txt", chapter_file)
if match:
chapter_num = int(match.group(1))
else:
print(f"Skipping file {chapter_file} as it does not match the expected format.")
continue
chapter_path = os.path.join(chapters_dir, chapter_file)
output_file_name = f"audio_chapter_{chapter_num}.wav"
output_file_path = os.path.join(output_audio_dir, output_file_name)
temp_audio_directory = os.path.join(".", "Working_files", "temp")
os.makedirs(temp_audio_directory, exist_ok=True)
temp_count = 0
with open(chapter_path, 'r', encoding='utf-8') as file:
chapter_text = file.read()
sentences = sent_tokenize(chapter_text, language='italian' if language == 'it' else 'english')
for sentence in tqdm(sentences, desc=f"Chapter {chapter_num}"):
fragments = split_long_sentence(sentence, max_length=249 if language == "en" else 213, max_pauses=10)
for fragment in fragments:
if fragment != "":
print(f"Generating fragment: {fragment}...")
fragment_file_path = os.path.join(temp_audio_directory, f"{temp_count}.wav")
if custom_model:
out = model.inference(fragment, language, gpt_cond_latent, speaker_embedding, temperature=0.7)
torchaudio.save(fragment_file_path, torch.tensor(out["wav"]).unsqueeze(0), 24000)
else:
speaker_wav_path = target_voice_path if target_voice_path else default_target_voice_path
language_code = language if language else default_language_code
tts.tts_to_file(text=fragment, file_path=fragment_file_path, speaker_wav=speaker_wav_path, language=language_code)
temp_count += 1
combine_wav_files(temp_audio_directory, output_audio_dir, output_file_name)
wipe_folder(temp_audio_directory)
print(f"Converted chapter {chapter_num} to audio.")
def convert_chapters_to_audio_standard_model(chapters_dir, output_audio_dir, target_voice_path=None, language=None):
selected_tts_model = "tts_models/multilingual/multi-dataset/xtts_v2"
tts = TTS(selected_tts_model, progress_bar=False).to(device)
if not os.path.exists(output_audio_dir):
os.makedirs(output_audio_dir)
for chapter_file in sorted(os.listdir(chapters_dir)):
if chapter_file.endswith('.txt'):
match = re.search(r"chapter_(\d+).txt", chapter_file)
if match:
chapter_num = int(match.group(1))
else:
print(f"Skipping file {chapter_file} as it does not match the expected format.")
continue
chapter_path = os.path.join(chapters_dir, chapter_file)
output_file_name = f"audio_chapter_{chapter_num}.wav"
output_file_path = os.path.join(output_audio_dir, output_file_name)
temp_audio_directory = os.path.join(".", "Working_files", "temp")
os.makedirs(temp_audio_directory, exist_ok=True)
temp_count = 0
with open(chapter_path, 'r', encoding='utf-8') as file:
chapter_text = file.read()
sentences = sent_tokenize(chapter_text, language='italian' if language == 'it' else 'english')
for sentence in tqdm(sentences, desc=f"Chapter {chapter_num}"):
fragments = split_long_sentence(sentence, max_length=249 if language == "en" else 213, max_pauses=10)
for fragment in fragments:
if fragment != "":
print(f"Generating fragment: {fragment}...")
fragment_file_path = os.path.join(temp_audio_directory, f"{temp_count}.wav")
speaker_wav_path = target_voice_path if target_voice_path else default_target_voice_path
language_code = language if language else default_language_code
tts.tts_to_file(text=fragment, file_path=fragment_file_path, speaker_wav=speaker_wav_path, language=language_code)
temp_count += 1
combine_wav_files(temp_audio_directory, output_audio_dir, output_file_name)
wipe_folder(temp_audio_directory)
print(f"Converted chapter {chapter_num} to audio.")
# Define the functions to be used in the Gradio interface
def convert_ebook_to_audio(ebook_file, target_voice_file, language, use_custom_model, custom_model_file, custom_config_file, custom_vocab_file, custom_model_url=None, progress=gr.Progress()):
ebook_file_path = ebook_file.name
target_voice = target_voice_file.name if target_voice_file else None
custom_model = None
working_files = os.path.join(".", "Working_files", "temp_ebook")
full_folder_working_files = os.path.join(".", "Working_files")
chapters_directory = os.path.join(".", "Working_files", "temp_ebook")
output_audio_directory = os.path.join(".", 'Chapter_wav_files')
remove_folder_with_contents(full_folder_working_files)
remove_folder_with_contents(output_audio_directory)
if use_custom_model and custom_model_file and custom_config_file and custom_vocab_file:
custom_model = {
'model': custom_model_file.name,
'config': custom_config_file.name,
'vocab': custom_vocab_file.name
}
if use_custom_model and custom_model_url:
print(f"Received custom model URL: {custom_model_url}")
download_dir = os.path.join(".", "Working_files", "custom_model")
download_and_extract_zip(custom_model_url, download_dir)
# Check if vocab.json exists and rename it
if rename_vocab_file_if_exists(download_dir):
print("vocab.json file was found and renamed.")
custom_model = {
'model': os.path.join(download_dir, 'model.pth'),
'config': os.path.join(download_dir, 'config.json'),
'vocab': os.path.join(download_dir, 'vocab.json_')
}
try:
progress(0, desc="Starting conversion")
except Exception as e:
print(f"Error updating progress: {e}")
if not calibre_installed():
return "Calibre is not installed."
try:
progress(0.1, desc="Creating chapter-labeled book")
except Exception as e:
print(f"Error updating progress: {e}")
create_chapter_labeled_book(ebook_file_path)
audiobook_output_path = os.path.join(".", "Audiobooks")
try:
progress(0.3, desc="Converting chapters to audio")
except Exception as e:
print(f"Error updating progress: {e}")
if use_custom_model:
convert_chapters_to_audio_custom_model(chapters_directory, output_audio_directory, target_voice, language, custom_model)
else:
convert_chapters_to_audio_standard_model(chapters_directory, output_audio_directory, target_voice, language)
try:
progress(0.9, desc="Creating M4B from chapters")
except Exception as e:
print(f"Error updating progress: {e}")
create_m4b_from_chapters(output_audio_directory, ebook_file_path, audiobook_output_path)
# Get the name of the created M4B file
m4b_filename = os.path.splitext(os.path.basename(ebook_file_path))[0] + '.m4b'
m4b_filepath = os.path.join(audiobook_output_path, m4b_filename)
try:
progress(1.0, desc="Conversion complete")
except Exception as e:
print(f"Error updating progress: {e}")
print(f"Audiobook created at {m4b_filepath}")
return f"Audiobook created at {m4b_filepath}", m4b_filepath
def list_audiobook_files(audiobook_folder):
# List all files in the audiobook folder
files = []
for filename in os.listdir(audiobook_folder):
if filename.endswith('.m4b'): # Adjust the file extension as needed
files.append(os.path.join(audiobook_folder, filename))
return files
def download_audiobooks():
audiobook_output_path = os.path.join(".", "Audiobooks")
return list_audiobook_files(audiobook_output_path)
# Gradio UI setup
def run_gradio_interface():
language_options = [
"en", "es", "fr", "de", "it", "pt", "pl", "tr", "ru", "nl", "cs", "ar", "zh-cn", "ja", "hu", "ko"
]
theme = gr.themes.Soft(
primary_hue="blue",
secondary_hue="blue",
neutral_hue="blue",
text_size=gr.themes.sizes.text_md,
)
with gr.Blocks(theme=theme) as demo:
gr.Markdown(
"""
# eBook to Audiobook Converter
Transform your eBooks into immersive audiobooks with optional custom TTS models.
This interface is based on [Ebook2AudioBookXTTS](https://github.com/DrewThomasson/ebook2audiobookXTTS).
"""
)
with gr.Row():
with gr.Column(scale=3):
ebook_file = gr.File(label="eBook File")
target_voice_file = gr.File(label="Target Voice File (Optional)")
language = gr.Dropdown(label="Language", choices=language_options, value="en")
with gr.Column(scale=3):
use_custom_model = gr.Checkbox(label="Use Custom Model")
custom_model_file = gr.File(label="Custom Model File (Optional)", visible=False)
custom_config_file = gr.File(label="Custom Config File (Optional)", visible=False)
custom_vocab_file = gr.File(label="Custom Vocab File (Optional)", visible=False)
custom_model_url = gr.Textbox(label="Custom Model Zip URL (Optional)", visible=False)
convert_btn = gr.Button("Convert to Audiobook", variant="primary")
output = gr.Textbox(label="Conversion Status")
audio_player = gr.Audio(label="Audiobook Player", type="filepath")
download_btn = gr.Button("Download Audiobook Files")
download_files = gr.File(label="Download Files", interactive=False)
convert_btn.click(
convert_ebook_to_audio,
inputs=[ebook_file, target_voice_file, language, use_custom_model, custom_model_file, custom_config_file, custom_vocab_file, custom_model_url],
outputs=[output, audio_player]
)
use_custom_model.change(
lambda x: [gr.update(visible=x)] * 4,
inputs=[use_custom_model],
outputs=[custom_model_file, custom_config_file, custom_vocab_file, custom_model_url]
)
download_btn.click(
download_audiobooks,
outputs=[download_files]
)
# Get the correct local IP or localhost
hostname = socket.gethostname()
local_ip = socket.gethostbyname(hostname)
# Ensure Gradio runs and prints the correct local IP
print(f"Running on local URL: http://{local_ip}:7860")
print(f"Running on local URL: http://localhost:7860")
# Launch Gradio app
demo.launch(server_name="0.0.0.0", server_port=7860, share=args.share)
# Check if running in headless mode
if args.headless:
if not args.ebook:
print("Error: In headless mode, you must specify an ebook file using --ebook.")
exit(1)
ebook_file_path = args.ebook
target_voice = args.voice if args.voice else None
custom_model = None
if args.use_custom_model:
custom_model = {
'model': args.custom_model,
'config': args.custom_config,
'vocab': args.custom_vocab
}
# Example headless execution
convert_ebook_to_audio(ebook_file_path, target_voice, args.language, args.use_custom_model, args.custom_model, args.custom_config, args.custom_vocab)
else:
# Launch Gradio UI
run_gradio_interface() |