Spaces:
Runtime error
Runtime error
A newer version of the Gradio SDK is available: 6.14.0
metadata
title: Russian Audiobook Studio
emoji: π§
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false
license: mit
short_description: Create Russian audiobooks from EPUB files with advanced TTS
π§ Russian Audiobook Studio
Transform your Russian EPUB books into professional-quality audiobooks using advanced Text-to-Speech technology with automatic accent placement and voice customization.
β¨ Features
- π EPUB Processing: Upload EPUB files and automatically extract chapters
- π― Chapter Selection: Choose specific chapters to process with checkbox interface
- π΅ Advanced TTS: High-quality Russian TTS with automatic accent placement using ruaccent
- ποΈ Voice Controls: Customize voice clarity, variation, and volume levels
- π§ Audio Preview: Built-in audio players for chapter previews
- π₯ Download Options: Download individual chapters or complete audiobook
- π± Responsive Design: Works on desktop and mobile devices
- βΏ Accessibility: Full keyboard navigation and screen reader support
π How to Use
- Upload EPUB: Click "Choose File" and select your Russian EPUB book
- Validate: Click "Validate EPUB" to extract chapters and book information
- Select Chapters: Choose which chapters to process using the checkbox interface
- Adjust Settings: Customize voice parameters in Advanced Settings
- Process: Click "Process Selected Chapters" to generate audiobook
- Preview & Download: Listen to previews and download your audiobook
βοΈ Advanced Settings
- Speech Speed: Adjust playback speed (0.6x to 1.4x)
- Quality Steps: Higher values = better quality, longer processing time
- Voice Clarity: Control how closely voice follows reference (1.0-4.0)
- Voice Variation: Control natural voice variation (-2.0 to 1.0)
- Volume Level: Target volume level for generated audio
π― Technical Features
- Automatic Accent Placement: Uses ruaccent library for perfect Russian pronunciation
- Professional Audio Quality: 24kHz, 16-bit, mono WAV output
- Real-time Progress: Live updates during processing with cancellation support
- Error Handling: Robust error handling with detailed status messages
- File Management: Automatic file serving and download functionality
π Requirements
- Russian EPUB files with proper text encoding
- Reference audio sample (6-12 seconds) for voice cloning
- Reference text (transcript of the reference audio)
π§ Technical Stack
- TTS Engine: ESpeech-TTS-1_RL-V2 (F5-TTS)
- Accent Processing: ruaccent for automatic Russian stress placement
- Audio Processing: librosa, soundfile, pydub
- Web Interface: Gradio with responsive design
- EPUB Processing: ebooklib for book parsing
π Notes
- Processing time depends on chapter length and quality settings
- Higher quality settings require more processing time
- All audio files are temporarily stored for preview and download
- The system automatically handles Russian text normalization and accent placement
π Ready to Create Audiobooks?
Upload your Russian EPUB file and start creating professional-quality audiobooks in minutes!