DNN-based source enhancement self-optimized by reinforcement learning using sound quality measurements
Authors:
- 1 March 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 81-85
Abstract
We investigated whether a deep neural network (DNN)-based source enhancement function can be self-optimized by reinforcement learning (RL). The use of a DNN is a powerful approach to describing the relationship between two sets of variables and can be useful for source enhancement function design. By training the DNN using a huge amount of training data, sound quality of output signals are improved. However, collecting a huge amount of training data is often difficult in practice. To use limited training data efficiently, we focus on the “self-optimization” of DNN-based source enhancement function in which RL is commonly utilized in the development of game playing computers. As a reward for RL, quantitative metrics that reflect a human's perceptual score (perceptual score), e.g., perceptual evaluation methods for audio source separation (PEASS), are utilized. To investigate whether the sound quality is improved by RL-based source enhancement, subjective tests were conducted. It was confirmed that the output sound quality of the RL-based source enhancement function improved as the number of iterations was increased and finally outperformed the conventional method.Keywords
This publication has 21 references indexed in Scilit:
- Phase-sensitive and recognition-boosted speech separation using deep recurrent neural networks2015
- Human-level control through deep reinforcement learningNature, 2015
- A Regression Approach to Speech Enhancement Based on Deep Neural NetworksIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- Ideal ratio mask estimation using deep neural networks for robust speech recognition2013
- Object-based audio for interactive football broadcastMultimedia Tools and Applications, 2013
- Frame-wise model re-estimation method based on Gaussian pruning with weight normalization for noise robust voice activity detectionSpeech Communication, 2012
- Temporal difference learning and TD-GammonCommunications of the ACM, 1995
- ATR Japanese speech database as a tool of speech recognition and synthesisSpeech Communication, 1990
- Learning representations by back-propagating errorsNature, 1986
- Speech enhancement using a minimum-mean square error short-time spectral amplitude estimatorIEEE Transactions on Acoustics, Speech, and Signal Processing, 1984