SpeechT5: Unified-Modal Encoder-Decoder Pre-Training for Spoken Language Processing
Paper
•
2110.07205
•
Published
•
5
SpeechT5
The SpeechT5 framework consists of a shared seq2seq and six modal-specific (speech/text) pre/post-nets that can address a few audio-related tasks.
-
microsoft/speecht5_tts
Text-to-Speech • Updated • 97.5k • 765Note Text-to-speech version of SpeechT5
-
219
SpeechT5 Speech Synthesis Demo
👩
microsoft/speecht5_vc
Audio-to-Audio • Updated • 2.11k • 104Note Voice-conversion version of SpeechT5
-
96
SpeechT5 Voice Conversion Demo
👩
microsoft/speecht5_asr
Automatic Speech Recognition • Updated • 2.67k • 41Note Automatic-speech-recognition version of SpeechT5
-
36
SpeechT5 Speech Recognition Demo
👩
microsoft/speecht5_hifigan
Updated • 88.9k • 20Note SpeechT5 produces a spectrogram, this model converts it to a waveform