yl4579/HiFTNet
HiFTNet: A Fast High-Quality Neural Vocoder with Harmonic-plus-Noise Filter and Inverse Short Time Fourier Transform
HiFTNet helps create high-quality, natural-sounding synthetic speech from raw speech features, like mel-spectrograms. It converts these abstract sound representations into actual audio waveforms, making digital voices sound more human-like. This is ideal for professionals developing applications that require realistic computer-generated speech, such as virtual assistants or audiobook narrators.
247 stars. No commits in the last 6 months.
Use this if you need to generate high-fidelity speech from mel-spectrograms quickly and efficiently, especially for real-time applications.
Not ideal if your primary goal is to extract speech features or perform voice analysis, as this tool focuses on speech synthesis.
Stars
247
Forks
23
Language
Python
License
MIT
Category
Last pushed
Jan 14, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/yl4579/HiFTNet"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
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