[NeurIPS 2021] Multiscale Benchmarks for Multimodal Representation Learning

The SpeechBrain project aims to build a novel speech toolkit fully based on PyTorch. With SpeechBrain users can easily create speech processing systems, ranging from speech recognition (both HMM/DNN and end-to-end), speaker recognition, speech enhancement, speech separation, multi-microphone speech processing, and many others.

Pytorch implementation of subband decomposition

A MATLAB implementation of CHiME4 baseline Beamformit

Book Library of P&W Studio

Butterfly Voice Changing Bowtie (BVCB), simple Sound regulator/simulator（PureData Implementation）

Web Speech API - Demo App

Low-complexity Real-time Neural Network for Blind Bandwidth Extension of Wideband Speech demo page

Realtime speech recognition based on Web Speech API

Speaker Diarization using Python, Flask and Html

Web Audio Speech Synthesis and Speech Recognition Framework

Explore NLP Topic Models Trained on Speeches Held by G20-Country Representatives at the UN General Assembly

In this Project I have used Speech Recognition tools to perform speech-to-text and text-to-speech tasks, and I leveraged pre-trained Transformers language models to give the bot some Artificial Intelligence.

Manage and schedule notes with Speech recognition feature

Your Personal AI Pronunciation Coach