Dockerfile for kaldi-gstreamer-server.

Scripts para treino de modelos acústicos

HHM-based Arabic ASR using Kaldi engine

Resources for easily building ASR systems with Kaldi

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

End-to-End Arabic ASR using DeepSpeech engine

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

This is now the official location of the Kaldi project.

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

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.

BurrMill core

Playing with kaldi, try make myown recipes.

Kaldi ASR Spanish example using the DIMEx100 corpus

This is now the official location of the Kaldi project.

Scripts e recursos para ASR em Português Brasileiro

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

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