# tensorflow/lingvo

Lingvo

Repository: https://github.com/tensorflow/lingvo
Canonical: https://ross.abutalabs.com/products/lingvo
Language: Python
License: Apache-2.0
License Family: permissive
Topics: speech-recognition, translation, speech-to-text, machine-translation, mnist, seq2seq, language-model, tts, asr, lm, nlp, tensorflow, speech, research, distributed, gpu-computing, speech-synthesis
Last push: 2026-06-22T19:13:51+00:00

## Health v2 (maintenance only)
Score: 72/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 88, release rhythm 35, longevity 100
- inputs: {"age_days": 2962, "days_push": 72, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2864, forks 451 (observed 2026-08-28T04:07:26.507371+00:00)

## What it is
Lingvo is a TensorFlow-based framework for building neural networks, particularly sequence models, with a focus on speech recognition, machine translation, language modeling, and TTS. It includes reference model implementations and supports distributed training on GPUs.

## Use cases
- train an automatic speech recognition model
- build a sequence-to-sequence model in TensorFlow
- run machine translation experiments
- train a large transformer language model with GShard
- do speech synthesis research
- run distributed training of sequence models on GPUs

## When to choose
- you need a research framework for speech or translation sequence models on TensorFlow
- you want reference implementations of ASR, MT, or LM models
- you need distributed training of large sequence models

## When to avoid
- you prefer PyTorch or JAX ecosystems
- you need production-ready speech APIs rather than a research framework
- you need frequent releases with backward-compatibility guarantees

## Facets
- artifact type: framework
- maturity: maintenance
- function: machine-learning, deep-learning, speech-recognition, tts, nlp, llm-training
- domain: machine-learning, speech-processing, deep-learning
- platform: python, cross-platform
- tags: tensorflow, sequence-models, asr, machine-translation, seq2seq, distributed-training, research, natural-language-processing, gpu, linux

## Member repositories
- tensorflow/lingvo (main) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:26.507371+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 2026-08-30T07:36:15.203958+00:00, confidence not recorded.
  - readme: https://github.com/tensorflow/lingvo (fetched 2026-08-28T04:07:26.507371+00:00, sha ad3c437493a8)
- Data as of 2026-08-30T08:39:29.467469+00:00.