Dockerfile for kaldi-gstreamer-server.

This is now the official location of the Kaldi project.

Some utility classes for using Kaldi in more flexible way.

The overarching problem that this project aims to address is the accessibility and effectiveness of education for university students, particularly international students, and those with disabilities. The challenge lies in the difficulties these students face in comprehending and retaining lecture content due to language barriers or disabilities

Practiced implementing a simple DNN (along with discriminative pretraining) for acoustic modelling for ASR and used kaldi for frontend and decode. Kaldi features, alignment, etc are used in python using Karel's (https://github.com/vesis84) library (https://github.com/vesis84/kaldi-io-for-python).

Use kaldi pretrained nnet3 model to align individual sentences and get phone-level transcripts

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

ToDoList - Project of HIT Industry Intern and Practice

Exploratory part of the implementation of the paper titled - "Improving Contextual Recognition in Automatic Speech Recognition Systems using Semantic Lattice Rescoring"

A Speech to Text Personal Assist inspired by kaldi2 and joint-bert

Neural network phone duration model on top of the Kaldi speech recognition framework

FastCGI support for Kaldi ASR

New compilation guide for Kaldi to WASM

Graduation project pipeline for Auto Speaker Verification system

AI 이노베이션 스퀘어 고급과정 - 음성지능 수업내용 (20190701-20190831 160시간)

Speech-to-text/quick chat for Rocket League via Kaldi

本仓库为深蓝学院-语音识别从入门到精通课程的作业仓库，仓库将随着课程的推进而更新。

Scripts used to do a experiment using kaldi for Sinhala Speech Recognition.

Fork from the official Kaldi project, and add with my comments here.