Speech enhancement for robust automatic speech recognition: Evaluation using a baseline system and instrumental measures

• Published in: Computer speech & language Vol. 46; pp. 574 - 584
• Main Authors: Moore, A.H., Peso Parada, P., Naylor, P.A.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-11-2017
• Subjects: Automatic speech recognition; Beamforming; Dereverberation; Microphone array signal processing; Realistic environments; Speech enhancement; Speech enhancement; Beamforming; Realistic environments; Microphone array signal processing; Dereverberation; Automatic speech recognition
• ISSN: 0885-2308; 1095-8363
• DOI: 10.1016/j.csl.2016.11.003

•Evaluation of baseline CHiME3 recogniser in diverse range of acoustic conditions.•Performance curves indicate relative influence of noise and reverberation.•Evaluation of 6 different speech enhancement pipelines.•Deverberation and beamforming dramatically improve performance in all conditions.•Improvement in STOI predicts improvement in WER.
1Present address: Cirrus Logic, Marble Arch House, 66 Seymour St., 1st Floor, London W1H 5BT, United Kingdom.Automatic speech recognition in everyday environments must be robust to significant levels of reverberation and noise. One strategy to achieve such robustness is multi-microphone speech enhancement. In this study, we present results of an evaluation of different speech enhancement pipelines using a state-of-the-art ASR system for a wide range of reverberation and noise conditions. The evaluation exploits the recently released ACE Challenge database which includes measured multichannel acoustic impulse responses from 7 different rooms with reverberation times ranging from 0.33 to 1.34 s. The reverberant speech is mixed with ambient, fan and babble noise recordings made with the same microphone setups in each of the rooms. In the first experiment, performance of the ASR without speech processing is evaluated. Results clearly indicate the deleterious effect of both noise and reverberation. In the second experiment, different speech enhancement pipelines are evaluated with relative word error rate reductions of up to 82%. Finally, the ability of selected instrumental metrics to predict ASR performance improvement is assessed. The best performing metric, Short-Time Objective Intelligibility Measure, is shown to have a Pearson correlation coefficient of 0.79, suggesting that it is a useful predictor of algorithm performance in these tests.