#152 · Primary category： Speech & Audio
PaddlePaddle-DeepSpeech
Speech recognition based on PaddlePaddle, supporting Chinese, with complete project and good accuracy; supports training and prediction on Windows and Linux, and prediction on Nvidia Jetson.
Project last updated：12/17/25
GitHub Stars
762
Forks
146
Contributors
2
License
Apache-2.0
Why we included this project
PaddlePaddle-DeepSpeech is a practical implementation of the DeepSpeech2 architecture for Mandarin speech recognition, built on PaddlePaddle. It walks through the whole workflow, from data preparation and augmentation to training, beam-search decoding, and model export, so a small team can go from raw audio to a working recognizer without stitching together scattered scripts. The author publishes pretrained models trained on AIShell, LibriSpeech, and WenetSpeech, which gives you a reasonable starting point for Chinese and English recognition instead of training from scratch. Deployment is covered as well: VAD-based long-audio segmentation, a web inference server, a GUI, and Nvidia Jetson support for edge hardware. If you want newer model options, the dynamic-graph successor PPASR adds Conformer and Squeezeformer.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
whisper.cpp
Port of OpenAI's Whisper model in C/C++
Real-Time-Voice-Cloning
Clone a voice in 5 seconds to generate arbitrary speech in real-time
VibeVoice
Open-Source Frontier Voice AI
voicebox
The open-source AI voice studio. Clone, dictate, create.
TTS
🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production