# PaddlePaddle/PaddleSpeech

Easy-to-use Speech Toolkit including Self-Supervised Learning model, SOTA/Streaming ASR with punctuation, Streaming TTS with text frontend, Speaker Verification System, End-to-End Speech Translation and Keyword Spotting. Won NAACL2022 Best Demo Award.

Repository: https://github.com/PaddlePaddle/PaddleSpeech
Canonical: https://ross.abutalabs.com/products/paddlespeech
Homepage: https://paddlespeech.readthedocs.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: transformer, conformer, speech-translation, streaming-asr, speech-alignment, punctuation-restoration, streaming-tts, speech-synthesis, tts, asr, kws, speech-recognition, sound-classification, voice-cloning, vocoder, voice-recognition, self-supervised-learning, wav2vec2, whisper, code-switch
Last push: 2026-08-12T01:07:51+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 8, longevity 100
- inputs: {"age_days": 3214, "days_push": 22, "days_rel": 546, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12670, forks 1960 (observed 2026-08-28T04:10:59.535895+00:00)

## What it is
PaddleSpeech is an open-source speech and audio toolkit built on the PaddlePaddle deep learning platform, covering ASR with punctuation, streaming TTS with a text frontend, speaker verification, speech translation, keyword spotting, and self-supervised speech models. It offers Python APIs, CLI commands, and a serving server, and won the NAACL 2022 Best Demo Award.

## Use cases
- transcribe speech audio to text with punctuation
- convert text to natural-sounding speech
- build a streaming speech recognition service
- verify speaker identity from voice samples
- translate spoken audio across languages end to end
- detect wake words with keyword spotting
- clone a voice from sample audio

## When to choose
- you want a single toolkit covering ASR, TTS, speaker verification, and speech translation
- you need streaming (real-time) ASR or TTS
- you prefer PaddlePaddle as your deep learning backend
- you want pretrained SOTA models with CLI and server deployment options

## When to avoid
- your stack is PyTorch-based and you want ecosystem compatibility (consider ESPnet or Torchaudio)
- you need only lightweight inference without the PaddlePaddle dependency
- you require extensive non-Chinese/English language coverage

## Facets
- artifact type: library
- maturity: active
- function: speech-recognition, tts, audio-processing, machine-learning, cli, http-server
- domain: speech-processing, machine-learning, deep-learning
- platform: python, windows, cli
- tags: asr, tts, speech-translation, speaker-verification, keyword-spotting, streaming, voice-cloning, paddlepaddle, vocoder, self-supervised-learning, natural-language-processing, linux, macos

## Member repositories
- PaddlePaddle/PaddleSpeech (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:59.535895+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 2026-08-29T17:13:52.679011+00:00, confidence not recorded.
  - readme: https://github.com/PaddlePaddle/PaddleSpeech (fetched 2026-08-28T04:10:59.535895+00:00, sha 0ba9e5ec7d4c)
  - homepage: https://paddlespeech.readthedocs.io (fetched 2026-08-29T08:10:11.721516+00:00, sha e841d4bfd6af)
  - registry_pypi: https://pypi.org/pypi/paddlespeech/json (fetched 2026-08-29T08:10:11.725393+00:00, sha d42d13472d32)
- Data as of 2026-08-30T08:39:29.467469+00:00.