PaddleSpeech: open-source toolkit for speech and audio
An open-source toolkit on the PaddlePaddle platform for speech and audio tasks; it won the NAACL2022 Best Demo Award.
Verified
Core facts
Read by a script from the official API, not rewritten by a model.
- License
- Apache-2.0
- Languages
- Python
- Maintainer
- PaddlePaddle
- Verified
What it is
PaddleSpeech is an open-source toolkit on the PaddlePaddle platform for a variety of critical tasks in speech and audio. PaddleSpeech won the NAACL2022 Best Demo Award. The Text-to-Speech module was originally called Parakeet and is now merged into the PaddleSpeech repository. PaddleSpeech is provided under the Apache-2.0 License.
Main features
Its Chinese frontend contains Text Normalization and Grapheme-to-Phoneme (G2P, including Polyphone and Tone Sandhi).
The Chinese frontend also uses self-defined linguistic rules to adapt Chinese context.
The toolkit contains audio functions like Automatic Speech Recognition, Text-to-Speech Synthesis, Speaker Verification, KeyWord Spotting, Audio Classification, and Speech Translation.
Getting started
There are two quick installation methods for PaddleSpeech: pip installation and source code compilation (recommended).
PaddleSpeech depends on paddlepaddle.
The PaddleSpeech models can be tried with the PaddleSpeech command line or Python.
The input option can be changed to test your own audio or text, and 16k wav format audio is supported.
The speech server can be tried with the PaddleSpeech Server command line.
The streaming ASR and streaming TTS server can also be tried.
pip install paddlepaddle -i https://mirror.baidu.com/pypi/simplepip install pytest-runner
pip install paddlespeechgit clone https://github.com/PaddlePaddle/PaddleSpeech.git
cd PaddleSpeech
pip install pytest-runner
pip install .
# If you need to install in editable mode, you need to use --use-pep517. The command is as follows:
# pip install -e . --use-pep517wget -c https://paddlespeech.cdn.bcebos.com/PaddleAudio/zh.wav
wget -c https://paddlespeech.cdn.bcebos.com/PaddleAudio/en.wavWho it is for
The authors strongly recommend installing PaddleSpeech in Linux with Python 3.8 or higher. The dependencies require gcc at least version 4.8.5. Supported operating systems are Linux (recommended), Windows and Mac OSX.
Speech translation uses pre-compiled Kaldi-related tools, and the speech translation demo can only be tried on Ubuntu. The sound classification model is based on 527 categories of the AudioSet dataset. The toolkit's vision is to empower industrial application and academic research.
The toolkit implements modules in the whole pipeline of speech tasks and uses mainstream datasets such as LibriSpeech, LJSpeech, AIShell and CSMSC. The toolkit also combines the workflows of these tasks with other fields such as Natural Language Processing (NLP) and Computer Vision (CV). Text-to-Speech in PaddleSpeech mainly contains three modules: Text Frontend, Acoustic Model and Vocoder.
On 2025.09.01, PaddleSpeech added the Whisper large v3 and turbo model. On 2023.02.16, PaddleSpeech added Cantonese TTS. Speech Translation is an end-to-end English to Chinese speech translation tool.
PaddleBoBo uses PaddleSpeech TTS to generate virtual human voice. VTuberTalk uses PaddleSpeech TTS and ASR to clone voice from videos.
Related tools
Related tools listed on this site.
百灵大模型
Ant Ling is a language and multimodal model family. Official documentation describes online multi-turn conversations, text generation/content creation, API access and cross-modal understanding/generation.
秒听
Tinrec is a meeting-notes and speech-to-text tool. Its public site describes live recordings, audio/video files and video-link transcription, followed by transcripts, meeting notes, speaker summaries and action items.
讯飞译制
iFlyrec Yizhi is a tool for audiovisual translation, AI dubbing and subtitle production, with timecode matching and output described on its official page.
Trancy
A reading and video assistant for language learning. Its site describes bilingual subtitles, selected-text and full-page translation, sentence analysis and personal learning materials.
通义听悟
The site calls it an AI assistant for work and study, built on the Qwen model and audio/video AI, for recording and reading audio/video content. Docs say individuals can use it free after phone sign-up; enterprises can integrate via API.
SoundView
Video-localization tool whose official summary describes translation, dubbing, speech generation, lip sync and subtitle processing.
魔音工坊
Moyin is an AI text-to-speech tool whose official page describes voiceovers for short videos, audiobooks and other narration uses.
讯飞智作
iFLYTEK Zhizuo is an audio/video content-creation platform. Its official site describes text-to-speech, digital-presenter videos and custom appearance/voice assets, including workflows starting from scripts or PPT slides.
逗哥配音
Douge is an AI voiceover tool for short-video production; its official summary also describes script extraction and vocal separation.
ElevenLabs
Text-to-speech and voice selection for narration, character audio, and spoken content.
琅琅配音
Langlang is an AI text-to-speech tool whose official summary describes multilingual voices and emotional styles for uses such as audiobooks and short videos.
1forAll
1forAll's official page (which also uses the name Voice-gen.ai) presents a unified platform for voice, image and video generation: text to speech, voice cloning, Excel or PDF to speech and video generation, pay-as-you-go with credits that, it says, do not expire.
Sources
This page is written from the official sources below. Open them to check the original text.
- GitHub READMEhttps://github.com/PaddlePaddle/PaddleSpeechFetched
Verified ·
This page is written by AI from the official sources above and checked automatically. This site has not tested it. Found a mistake? Tell us.