Abstract : My talk will focus on robustness to background noise in distant-microphone speech recordings. I will introduce deep learning based techniques for speech enhancement and for acoustic modeling of speech. I will then report the results of a study on the impact of environment and microphone mismatches on the recognition accuracy. This study reveals that mismatched training data can sometimes outperform matched data. I will suggest a way to optimize the training set in order to exploit this finding.
https://hal.inria.fr/hal-01588876
Contributor : Emmanuel Vincent
<>
Submitted on : Sunday, September 17, 2017 - 8:56:33 PM
Last modification on : Wednesday, April 3, 2019 - 1:22:58 AM
Long-term archiving on : Monday, December 18, 2017 - 1:06:08 PM
Emmanuel Vincent. When mismatched training data outperform matched data. Systematic approaches to deep learning methods for audio, Sep 2017, Vienna, Austria. ⟨hal-01588876⟩
