U

Unit Hifigan Mhubert Vp En Es Fr It3 400k Layer11 Km1000 Lj Dur

Developed by facebook
A speech-to-speech translation model based on fairseq S2UT, supporting Spanish to English translation
Downloads 27
Release Time : 8/31/2022

Model Overview

This model is a discrete unit-based speech-to-speech translation model that uses the HiFiGAN vocoder for speech synthesis and supports translation from Spanish to English.

Model Features

Multilingual Support
Supports speech translation from Spanish to English
High-quality Speech Synthesis
Uses the HiFiGAN vocoder to generate high-quality speech output
Discrete Unit-based
Employs discrete unit representation for speech translation, improving efficiency and accuracy

Model Capabilities

Speech-to-speech translation
Text-to-speech
Multilingual processing

Use Cases

Speech Translation
Real-time Speech Translation
Translates Spanish speech into English speech in real-time
High-quality English speech output
Speech Synthesis
Speech Content Generation
Converts text content into natural speech
Natural and fluent speech output
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase