Vixtts
viⓍTTS is a voice generation model supporting 18 languages, specifically optimized for Vietnamese, achieving cross-lingual voice cloning with just 6 seconds of audio.
Downloads 2,782
Release Time : 4/4/2024
Model Overview
A voice synthesis model fine-tuned based on XTTS-v2.0.3, extended with Vietnamese tokenizer and trained on the viVoice dataset, supporting multilingual voice cloning.
Model Features
Cross-lingual Voice Cloning
Achieves voice conversion across different languages with just 6 seconds of audio sample
Vietnamese Optimization
Specifically extended Vietnamese tokenizer and fine-tuned on Vietnamese datasets
Multilingual Support
Supports voice synthesis in 18 languages
Model Capabilities
Text-to-Speech
Voice Cloning
Cross-lingual Voice Synthesis
Use Cases
Voice Synthesis
Multilingual Voice Assistant
Provides natural voice output for users in different languages
Supports fluent speech in 18 languages
Voice Cloning Application
Clones a specific speaker's voice based on short audio samples
Achieves voice conversion with just 6-second samples
Education
Language Learning Tool
Generates pronunciation demonstrations in different languages
Helps learners master correct pronunciation
Featured Recommended AI Models