# SeanNaren/deepspeech.pytorch

Speech Recognition using DeepSpeech2.

Repository: https://github.com/SeanNaren/deepspeech.pytorch
Canonical: https://ross.abutalabs.com/products/deepspeechpytorch
Language: Python
License: MIT
License Family: permissive
Last push: 2022-12-13T15:05:51+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3522, "days_push": 1359, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2136, forks 618 (observed 2026-08-28T04:06:17.625939+00:00)

## What it is
A PyTorch implementation of the DeepSpeech2 speech recognition model, built on PyTorch Lightning, supporting training, testing, and inference. It includes dataset preparation scripts for common speech corpora and optional beam-search decoding with a kenlm language model.

## Use cases
- train a speech-to-text model in pytorch
- transcribe audio with deepspeech2
- fine-tune asr on custom dataset
- run speech recognition inference with language model decoding
- train on librispeech or common voice

## When to choose
- you want a proven DeepSpeech2 implementation in PyTorch
- you need training scripts for standard ASR datasets like LibriSpeech or Common Voice
- you want optional kenlm beam-search decoding

## When to avoid
- you need state-of-the-art ASR accuracy from modern architectures
- you want an actively developed project with frequent updates
- you need a production-ready hosted speech API

## Facets
- artifact type: library
- maturity: maintenance
- function: speech-recognition, machine-learning, deep-learning, audio-processing
- domain: speech-processing, machine-learning, deep-learning
- platform: python
- tags: pytorch, deepspeech2, asr, pytorch-lightning, ctc, speech-to-text, kenlm, linux, docker, gpu

## Member repositories
- SeanNaren/deepspeech.pytorch (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:17.625939+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 2026-08-30T02:51:51.763914+00:00, confidence not recorded.
  - readme: https://github.com/SeanNaren/deepspeech.pytorch (fetched 2026-08-28T04:06:17.625939+00:00, sha bc0213f675ce)
- Data as of 2026-08-30T08:39:29.467469+00:00.