*DTW: Dynamic Time Warping

• Folders
  • unittest: python framework "unittest" learning. Run the file "run.py" with Python3.
  • sounds: sound files of "MFCC_DTW.ipynb".
  • train: train sample audios of "DTW_MFCC_KNN.ipynb".
  • test: test sample audio of "DTW_MFCC_KNN.ipynb".
• Files
  • wavToTag.txt: 245 French words.
  • dtw.py: implementation of DTW algorithm.
  • dtwTest.py: test of DTW algorithm.
  • VoiceCommand.py: a simple voice command recognition demo using DTW.
  • DTW_simple_example.ipynb: DTW simple example.
  • MFCC_DTW.ipynb: compare the MFCCs of two sounds using DTW.
  • speech_recognition.ipynb: simple speech recognition system can be implemented using DTW + MFCC.
  • DTW_MFCC_KNN.ipynb: simple speech recognition using DTW, MFCC, and kNN (k-NearestNeighbor)

• PyAudio
• Anaconda3 (Python3.5+)
• Librosa: pip install librosa

• https://github.com/pierre-rouanet/dtw
• https://github.com/slaypni/fastdtw
• https://github.com/psbots/dtwSpeechRecognition
• https://www.cnblogs.com/rockyf/articles/4519352.html
• https://blog.csdn.net/raym0ndkwan/article/details/45614813
• https://blog.csdn.net/zouxy09/article/details/9140207
• https://www.cnblogs.com/chuxiuhong/p/6124459.html
• https://www.cnblogs.com/51kata/p/5887940.html