Voice controlled electronic devices have become a popular trend. The advantage is that electronic devices can be controlled without hands. For this scheme, we use Google TensorFlow as the algorithm development environment of speech recognition for deep learning. Then we implement speech recognition on NuMaker-PFM-M487 platform, and realize an offline and instant speech recognition system by Keyword Spotting.

A complete deep learning speech recognition system requires two platforms. As Fig.1-1 shows, one is PC platform. We program the deep learning code and train the model by Tensorflow and Python. Due to the supervised learning (Note 1) for the training mode, it is necessary to give the system a large amount of training data and labels. Then extract the features of speech data and train the model by deep neural networks (DNN). Until the system reaches the optimization, we evaluate the accuracy by modifying the training model repeatedly. The other platform is NuMaker-PFM-M487. The speech recognition system can be implemented based on the training parameters from PC platform.

• Related IC/Platform
  • M487
  • NuMaker-PFM-M487
  • NuMaker emWin M487

*Note: Nuvoton and NuMicro are trademarks or registered trademarks of Nuvoton Technology Corporation. All other trademarks and copyrights mentioned herein are the property of their respective owners.