shivammehta25/Neural-HMM
Neural HMMs are all you need (for high-quality attention-free TTS)
This project helps speech synthesis researchers and developers create high-quality, natural-sounding synthetic speech from written text. It takes text input and generates corresponding audio waveforms, specifically for applications where attention-based models might be too complex or slow. The primary users are researchers or engineers working on text-to-speech (TTS) systems who need advanced, efficient speech generation models.
164 stars.
Use this if you are a researcher or engineer looking to develop or experiment with high-quality, attention-free text-to-speech models using neural Hidden Markov Models.
Not ideal if you need a plug-and-play solution for simple audio generation without deep technical involvement or if your primary goal is real-time, low-latency deployment on highly constrained devices.
Stars
164
Forks
26
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Mar 09, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/shivammehta25/Neural-HMM"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
yeyupiaoling/MASR
Pytorch实现的流式与非流式的自动语音识别框架，同时兼容在线和离线识别，目前支持Conformer、Squeezeformer、DeepSpeech2模型，支持多种数据增强方法。
shivammehta25/Matcha-TTS
[ICASSP 2024] 🍵 Matcha-TTS: A fast TTS architecture with conditional flow matching
coqui-ai/TTS
🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production
DigitalPhonetics/IMS-Toucan
Controllable and fast Text-to-Speech for over 7000 languages!
gabrielmittag/NISQA
NISQA - Non-Intrusive Speech Quality and TTS Naturalness Assessment