We describe the structure and application of an acoustic room simulator to generate large-scale simulated data for training deep neural networks for far-field speech recognition. The system simulates millions of different room dimensions, a wide distribution of reverberation time and signal-to-noise ratios, and a range of microphone and sound source locations. We start with a relatively clean training set as the source and artificially create simulated data by randomly sampling a noise configuration for every new training example. As a result, the acoustic model is trained using examples that are virtually never repeated. We evaluate performance of this approach based on room simulation using a factored complex Fast Fourier Transform (CFFT) acoustic model introduced in our earlier work, which uses CFFT layers and LSTM AMs for joint multichannel processing and acoustic modeling. Results show that the simulator-driven approach is quite effective in obtaining large improvements not only in simulated test conditions, but also in real / rerecorded conditions. This room simulation system has been employed in training acoustic models including the ones for the recently released Google Home.
Cite as: Kim, C., Misra, A., Chin, K., Hughes, T., Narayanan, A., Sainath, T.N., Bacchiani, M. (2017) Generation of Large-Scale Simulated Utterances in Virtual Rooms to Train Deep-Neural Networks for Far-Field Speech Recognition in Google Home. Proc. Interspeech 2017, 379-383, doi: 10.21437/Interspeech.2017-1510
@inproceedings{kim17_interspeech,
author={Chanwoo Kim and Ananya Misra and Kean Chin and Thad Hughes and Arun Narayanan and Tara N. Sainath and Michiel Bacchiani},
title={{Generation of Large-Scale Simulated Utterances in Virtual Rooms to Train Deep-Neural Networks for Far-Field Speech Recognition in Google Home}},
year=2017,
booktitle={Proc. Interspeech 2017},
pages={379--383},
doi={10.21437/Interspeech.2017-1510}
}