# wenet-e2e/wenet

Production First and Production Ready End-to-End Speech Recognition Toolkit

Repository: https://github.com/wenet-e2e/wenet
Canonical: https://ross.abutalabs.com/products/wenet
Homepage: https://wenet-e2e.github.io/wenet/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: e2e-models, pytorch, asr, transformer, conformer, production-ready, automatic-speech-recognition, speech-recognition, whisper
Last push: 2026-06-15T05:56:43+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 87, release rhythm 8, longevity 100
- inputs: {"age_days": 2115, "days_push": 79, "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 5227, forks 1187 (observed 2026-08-28T04:09:13.278939+00:00)

## What it is
WeNet is a production-first, end-to-end automatic speech recognition (ASR) toolkit built on PyTorch with transformer/conformer models. It provides a Python package and CLI for transcription with pretrained models (Paraformer, Whisper, FireRed), plus training and deployment runtimes.

## Use cases
- transcribe audio files to text
- build a speech-to-text service in production
- train a custom ASR model on my own dataset
- run whisper or paraformer speech recognition from the command line
- deploy speech recognition on device or server with a lightweight runtime
- fine-tune an end-to-end ASR model with PyTorch

## When to choose
- you need production-ready, deployable speech recognition with runtime support
- you want pretrained Chinese and English ASR models with a simple Python/CLI API
- you want to train or fine-tune transformer/conformer ASR models on public or custom datasets

## When to avoid
- you need text-to-speech or other speech tasks beyond recognition
- you want a no-training, plug-and-play cloud transcription API instead of a self-hosted toolkit
- you work outside Python/C++ runtimes or lack GPU/CUDA for training

## Facets
- artifact type: library
- maturity: active
- function: speech-recognition, machine-learning, deep-learning, llm-inference, cli
- domain: speech-processing, machine-learning, developer-tools
- platform: python, cli, cross-platform
- tags: asr, pytorch, transformer, conformer, whisper, paraformer, end-to-end, production-ready, speech-to-text, transcription, natural-language-processing, linux, macos, gpu

## Member repositories
- wenet-e2e/wenet (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:13.278939+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 2026-08-29T17:59:40.834936+00:00, confidence not recorded.
  - readme: https://github.com/wenet-e2e/wenet (fetched 2026-08-28T04:09:13.278939+00:00, sha bec0f8af88f9)
  - homepage: https://wenet-e2e.github.io/wenet/ (fetched 2026-08-29T08:54:59.458971+00:00, sha ba3deb6a250a)
- Data as of 2026-08-30T08:39:29.467469+00:00.