Acoustic model

Hi, after reading Deepspeech.py, I’m feeling confuse. Is that acoustic model in DeepSpeech using Hidden Markov Model (HMM) or Gaussian Mixture Model (GMM) like many traditional ways?. If not, what did Deepspeech use to replace them ?

You can get more details on how DeepSpeech works from the original DeepSpeech paper by Baidu: https://arxiv.org/abs/1412.5567

1 Like