# mravanelli/pytorch-kaldi

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.

Repository: https://github.com/mravanelli/pytorch-kaldi
Canonical: https://ross.abutalabs.com/products/pytorch-kaldi
Language: Python
License Family: other
Topics: speech-recognition, gru, dnn, kaldi, rnn-model, pytorch, timit, deep-learning, deep-neural-networks, recurrent-neural-networks, multilayer-perceptron-network, lstm, lstm-neural-networks, speech, asr, rnn, dnn-hmm
Last push: 2022-03-14T05:10:47+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3109, "days_push": 1633, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2404, forks 442 (observed 2026-08-28T04:06:43.976006+00:00)

## What it is
PyTorch-Kaldi is a toolkit for developing state-of-the-art DNN/HMM hybrid speech recognition systems, combining PyTorch-managed neural networks with Kaldi-based feature extraction and decoding. It is now in maintenance mode, with the authors recommending migration to SpeechBrain.

## Use cases
- train dnn-hmm speech recognition models
- run asr experiments on timit
- build rnn/lstm acoustic models with pytorch
- hybrid kaldi and pytorch speech recognition pipeline
- research automatic speech recognition

## When to choose
- you need the classic DNN/HMM hybrid ASR setup with Kaldi decoding
- you are reproducing published PyTorch-Kaldi research results

## When to avoid
- you are starting a new speech project - use SpeechBrain instead
- you need end-to-end ASR or modern speech tasks like enhancement or separation

## Facets
- artifact type: library
- maturity: maintenance
- function: speech-recognition, machine-learning, deep-learning, audio-processing
- domain: speech-processing, machine-learning, deep-learning
- platform: python, cli
- tags: asr, kaldi, pytorch, dnn-hmm, rnn, lstm, speechbrain-successor, linux

## Member repositories
- mravanelli/pytorch-kaldi (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:43.976006+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:34:11.014977+00:00, confidence not recorded.
  - readme: https://github.com/mravanelli/pytorch-kaldi (fetched 2026-08-28T04:06:43.976006+00:00, sha c8cad207588c)
- Data as of 2026-08-30T08:39:29.467469+00:00.