wavernn

WaveRNN Vocoder + TTS

Phoneme multilingual(Russian-English) voice cloning based on

A PyTorch implementation of "Robust Universal Neural Vocoding"

Linear Prediction Coefficients estimation from mel-spectrogram implemented in Python based on Levinson-Durbin algorithm.

an improved version of Real-time-voice-cloning

Framework for one-shot multispeaker system based on Deep Learning

end-to-end voicebot that answers open domain questions.

Audio generation using WaveRNN and WaveNet

The (unofficial) vanilla version of WaveRNN

implementation of Nal et al.「Efficient Neural Audio Synthesis」

Image Reader : Getting voice file from Image.

🐙 Waver 1.0 is a universal model for T2V, I2V, and T2I generation, delivering 1080p output with flexible length and aspect ratios.