# Delta-ML/delta

DELTA is a deep learning based natural language and speech processing platform. LF AI & DATA Projects: https://lfaidata.foundation/projects/delta/

Repository: https://github.com/Delta-ML/delta
Canonical: https://ross.abutalabs.com/products/delta-ml-delta
Homepage: https://delta-didi.readthedocs.io/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: nlp, deep-learning, tensorflow, speech, sequence-to-sequence, seq2seq, speech-recognition, text-classification, speaker-verification, nlu, text-generation, emotion-recognition, tensorflow-serving, tensorflow-lite, inference, asr, serving, front-end, custom-ops, ops
Archived: true
Last push: 2025-04-16T19:27:35+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 16, release rhythm 8, longevity 100
- inputs: {"age_days": 2653, "days_push": 504, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1607, forks 282 (observed 2026-08-28T04:05:10.645849+00:00)

## What it is
DELTA is a deep learning based end-to-end natural language and speech processing platform built on TensorFlow and Python 3. It provides one-command training, configuration-driven tuning, and integrated serving for NLP and speech models.

## Use cases
- train text classification models
- build speech recognition (ASR) systems
- speaker verification models
- emotion recognition from speech
- named entity recognition and question answering
- deploy NLP models with TensorFlow Serving
- convert models to TensorFlow Lite for mobile inference

## When to choose
- you need an end-to-end TensorFlow pipeline for NLP or speech tasks
- you want training and serving with consistent data processing in the model graph
- you prefer config-file-driven model experimentation

## When to avoid
- you work primarily in PyTorch or JAX
- you need the latest transformer LLM tooling rather than classic NLP/speech tasks
- you need a very actively developed project with frequent updates

## Facets
- artifact type: framework
- maturity: maintenance
- function: machine-learning, deep-learning, nlp, speech-recognition, llm-training, llm-inference
- domain: speech-processing, machine-learning, deep-learning
- platform: python
- tags: tensorflow, seq2seq, asr, text-classification, speaker-verification, tensorflow-serving, tensorflow-lite, model-deployment, natural-language-processing, linux, docker

## Member repositories
- Delta-ML/delta (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:10.645849+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 2026-08-30T03:51:35.762944+00:00, confidence not recorded.
  - readme: https://github.com/Delta-ML/delta (fetched 2026-08-28T04:05:10.645849+00:00, sha 246a52bc007b)
- Data as of 2026-08-30T08:39:29.467469+00:00.