Shypd.aiThe largest open index of AI toolsFastSpeech2 is an open-source Audio & Music tool that provides a PyTorch implementation of Microsoft's FastSpeech 2 for fast and high-quality end-to-end text-to-speech. It supports multi-speaker English and Mandarin TTS.
Trending
About
FastSpeech2 offers a PyTorch implementation of Microsoft's "FastSpeech 2: Fast and High-Quality End-to-End Text to Speech" system. Building upon an existing FastSpeech implementation, this project facilitates research and development in speech synthesis. Key features include support for multi-speaker English and Mandarin text-to-speech, with models trained on datasets like LJSpeech, AISHELL-3, and LibriTTS. Users can control pitch, volume, and speaking rate of synthesized utterances. The implementation also integrates vocoders like MelGAN and HiFi-GAN, and provides scripts for preprocessing, training, and inference, making it a comprehensive solution for advanced TTS applications.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
