This repository is the official implementation of unimodal aggregation (UMA) for automaticspeech recognition (ASR).

A quick & dirty script to generate and view subtitles and transcriptions for your multimedia files using ggerganov/whisper.cpp

A Kaldi recipe for training a hybrid DNN-HMM speech recognition model

This repository contains scripts to prune Wav2vec2 using a neuroevolution-based method. More details about this method can be found in the paper Compressing Wav2vec2 for Embedded Applications.

Kaldi-based Korean ASR (한국어 음성인식) open-source project

Baseline Recipe for VoicePrivacy Challenge 2020: https://www.voiceprivacychallenge.org/vp2020/docs/VoicePrivacy_2020_Eval_Plan_v1_3.pdf

EC499: Major Project

Generate automated German subtitles using one of the three implemented machine learning generated language model.

HMM-based ASR systems trained on CommonVoice(zh-TW) using Kaldi.

A collection of various BASH scripts organised into subcategories. Most of these scripts were developed as assignments for the university's Systems & Networks Administration course. Others were developed to handle routine day-to-day tasks.

To evaluate OpenAI's Whisper library for transcribing audio into text

An Indian English ASR system based on Hidden Markov Models (HMM) has been designed using Kaldi(Povey et al., 2011).

☕🇧🇷 Scripts para o Kaldi em Português Brasileiro

Vosk ASR Docker images with GPU for Jetson boards, PCs, M1 laptops and GPC

Automated Speech Recognition

Kaldi-based audio-visual speech recognition

HHM-based Arabic ASR using Kaldi engine

End-to-End Arabic ASR using DeepSpeech engine

BurrMill core