Dubverse offers two powerful Text-to-Speech (TTS) models: neo.one and candy.two. Each model has its unique strengths and is designed for specific use cases. This guide will help you understand the features and applications of both models.
Feature | neo.one | candy.two | swift.call |
---|---|---|---|
Processing Speed | Ultra-fast | Standard | Real-time |
Voice Quality | Good quality | Highest fidelity | High quality |
Resource Usage | Optimized (lower) | Standard | Optimized |
Language Support | 10+ languages | 8+ languages | Hindi & English |
Emotion Control | Basic | Advanced | Basic |
Batch Processing | Optimized | Supported | Streaming |
Architecture | Non-autoregressive | Transformer-based | Voicebot-optimized |
Use Case | Efficient, large-scale TTS | High-quality, premium TTS | Real-time voicebot TTS |
Voice Cloning | No | No | Yes |
Word Timestamps | No | No | Yes |