pytorch-kaldi is 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.

deep-neural-networks deep-learning speech dnn pytorch recurrent-neural-networks lstm gru speech-recognition rnn kaldi rnn-model asr lstm-neural-networks multilayer-perceptron-network timit dnn-hmm

• Updated Mar 14, 2022
• Python

SincNet is a neural architecture for efficiently processing raw audio samples.

audio python deep-learning signal-processing waveform cnn pytorch artificial-intelligence speech-recognition neural-networks convolutional-neural-networks digital-signal-processing filtering speaker-recognition speaker-verification speech-processing audio-processing asr timit speaker-identification

• Updated Apr 28, 2021
• Python

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.

deep-learning neural-network speech speech-recognition neural-networks deeplearning speech-to-text speaker-recognition speaker-verification speech-processing speech-recognizer beamforming speech-analysis timit speechrecognition speech-api speech-separation librispeech speech-emotion-recognition speaker-identification

• Updated Jun 18, 2025
• HTML

The DARPA TIMIT Acoustic-Phonetic Continuous Speech Corpus.

speech timit-dataset timit darpa

• Updated Mar 5, 2022

End-to-End speech recognition implementation base on TensorFlow (CTC, Attention, and MTL training)

tensorflow end-to-end speech-recognition beam-search automatic-speech-recognition speech-to-text attention-mechanism asr timit-dataset ctc timit end-to-end-learning csj librispeech joint-ctc-attention

• Updated Jan 23, 2018
• Python

CTC end -to-end ASR for timit and 863 corpus.

decoder pytorch kaldi ctc timit

• Updated Dec 20, 2019
• Python

MXNet implementation of RNN Transducer (Graves 2012): Sequence Transduction with Recurrent Neural Networks

mxnet end-to-end speech-recognition transducers asr timit rnnt-joint rnn-transducer sequence-transduction rnnt-model

• Updated Jun 7, 2021
• Python

Tensorflow implementation of "Listen, Attend and Spell" authored by William Chan. This project utilizes input pipeline and estimator API of Tensorflow, which makes the training and evaluation truly end-to-end.

tensorflow speech-recognition seq2seq speech-to-text timit listen-attend-and-spell

• Updated Jan 31, 2019
• Python

Keras (tensorflow) implementation of SincNet (Mirco Ravanelli, Yoshua Bengio - https://github.com/mravanelli/SincNet)

audio machine-learning deep-learning neural-network tensorflow waveform keras cnn artificial-intelligence speech-recognition convolutional-neural-networks digital-signal-processing filtering speaker-recognition speaker-verification speech-processing audio-processing asr timit

• Updated May 18, 2021
• Python

Python implementation of pre-processing for End-to-End speech recognition

end-to-end dataset speech-recognition automatic-speech-recognition preprocessing transcription attention-mechanism timit-dataset ctc switchboard timit csj librispeech

• Updated Feb 19, 2018
• Python

Speech recognition on the TIMIT (or any other) dataset

theano neural-network speech speech-recognition phonemes timit

• Updated Nov 2, 2017
• Python

This code implements a basic MLP for speech recognition. The MLP is trained with pytorch, while feature extraction, alignments, and decoding are performed with Kaldi. The current implementation supports dropout and batch normalization. An example for phoneme recognition using the standard TIMIT dataset is provided.

python deep-neural-networks deep-learning cuda pytorch speech-recognition neural-networks feedforward-neural-network mlp kaldi asr timit multilayer-perceptron kaldi-asr

• Updated Feb 10, 2018
• Perl

Pytorch based phoneme recognition (TIMIT phoneme classification)

paper pytorch timit speechrecognition minimalgru cbhg

• Updated Apr 25, 2018
• Python

THEANO-KALDI-RNNs is a project implementing various Recurrent Neural Networks (RNNs) for RNN-HMM speech recognition. The Theano Code is coupled with the Kaldi decoder.

deep-neural-networks theano deep-learning recurrent-neural-networks gru rnn kaldi timit gated-recurrent-units theano-kaldi-rnns

• Updated Apr 15, 2018
• Perl

A Simple Automatic Speech Recognition (ASR) Model in Tensorflow, which only needs to focus on Deep Neural Network. It's easy to test popular cells (most are LSTM and its variants) and models (unidirectioanl RNN, bidirectional RNN, ResNet and so on). Moreover, you are welcome to play with self-defined cells or models.

deep-learning tensorflow lstm rnn automatic-speech-recognition ctc timit acoustic-model

• Updated Jan 18, 2018
• Python

Attention-based end-to-end ASR on TIMIT in PyTorch

end-to-end pytorch asr timit attention-seq2seq

• Updated Nov 9, 2021
• Python

Extract mfcc vectors and phones from TIMIT dataset

deep-learning phone speech-recognition data-preprocessing mfcc timit-dataset timit

• Updated Mar 23, 2023
• Shell

Sum-Product Networks (SPNs) for Robust Automatic Speaker Identification.

missing-data speaker-verification timit-dataset robustness timit sum-product-networks speaker-identification deep-xi spn-speaker-model robust-speaker-recognition ideal-binary-mask robust-speaker-identification robust-speaker-verification missing-feature-theory marginalisation marginalization

• Updated Aug 30, 2020
• Python

Python/numpy/pandas convenience wrapper for the TIMIT database.

audio python transcription audio-recordings phonemes timit timit-database phoneme-transcriptions timit-utils

• Updated Nov 26, 2018
• Jupyter Notebook

A collection of benchmarking tools.

benchmarking benchmark timer timit

• Updated Oct 13, 2026
• Python