MOSS-TTS Aims for More Robust Speech Synthesis
A new text-to-speech model introduces 'delay-pattern decoding' to solve common word skipping and repetition errors in parallel generation.
The OpenMOSS team has released MOSS-TTS v1.5, a new model for multilingual text-to-speech synthesis. Built on the VITS architecture, the model is designed to generate natural-sounding speech in Chinese, English, and Japanese.
Many modern TTS systems use parallel decoding to generate audio quickly, but this approach can lead to instability, causing the model to occasionally skip or repeat words. This lack of reliability is a significant hurdle for using such models in production applications.
To address this, MOSS-TTS introduces a technique its creators call "delay-pattern decoding." This novel mechanism is designed to mitigate common generation errors, significantly improving the model's robustness and making the output more consistent and reliable.
The model is available on the Hugging Face Hub for community use. While the code is licensed under the permissive MIT license, the model weights themselves are restricted to non-commercial research purposes.
Sources
- Visit
OpenMOSS-Team/MOSS-TTS-v1.5
Hugging Face
More in Text → Speech
StepFun's StepAudio 3 Realtime targets live voice AI
The audio-language foundation model builds a listen-converse-think-act loop aimed at natural, low-latency spoken interaction.
StepFun's StepAudio 3 Gen Unifies TTS and Music
A single discrete autoregressive model handles speech, voice design, sound effects, and music generation.
Breeze-TTS-2 Brings Open Voice Cloning to English
BreezeBlue's second-generation text-to-speech model pairs voice cloning with controllable direction, all under an open release on Hugging Face.
0 comments
No comments yet. Be the first to weigh in.