Nevis is a (sort of) all in one speech transcription library backed by Kaldi ASR.

Kaldi Implementation with Flask

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

Some of the kaldi functions with python wrappers.

Extract phone-level alignment and phonemic transcript from kaldi ali.*.gz files

This is a fork of PyTorch-Kaldi, a project for developing state-of-the-art DNN/RNN hybrid speech recognition systems. The DNN part is managed by pytorch, while feature extraction, label computation, and decoding are performed with the kaldi toolkit. This repo adds support to use a GAN front-end for an ASR acoustic model.

ASR-WebUI : Deploying kaldi Model to Azure

Playing with kaldi, try make myown recipes.

A kaldi recipe for modern standard arabic speech recognition

Kaldi ASR Architeture Proposal

Tool to transform data from Nemo/Deepspeech format to Kaldi as described here — https://kaldi-asr.org/doc/data_prep.html

Scripts run on Kaldi toolkit on Kaggle's Tensorflow Speech Recognition Challenge

Web application for speech recognition that can be configured with kaldi.

Aligner with MFA for Samromur dataset

Connect BigBlueButton conferences to Kaldi automatic-speech-recognition

A Kaldi-Recipe for Egyptian Arabic Speech Recognition

🔥 Deep NN Models with FRAMEWORKS -- PyTorch, Tensorflow 2, Kaldi, FastSpeech, MxNET Frameworks -- Image/Speech/NLP/Recommendation/Transformer Cognitive Analytics

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

My documentation of which script does what in Kaldi