Dereverberation of autoregressive envelopes for far-field speech recognition

Abstract

The task of speech recognition in far-field environments is adversely affected by the reverberant artifacts that elicit as the temporal smearing of the sub-band envelopes. In this paper, we develop a neural model for speech dereverberation using the long-term sub-band envelopes of speech. The sub-band envelopes are derived using frequency domain linear prediction (FDLP) which performs an autoregressive estimation of the Hilbert envelopes. The neural dereverberation model estimates the envelope gain which when applied to reverberant signals suppresses the late reflection components in the far-field signal. The dereverberated envelopes are used for feature extraction in speech recognition. Further, the sequence of steps involved in envelope dereverberation, feature extraction and acoustic modeling for ASR can be implemented as a single neural processing pipeline which allows the joint learning of the dereverberation network and the acoustic model. Several experiments are performed on the REVERB challenge dataset, CHiME-3 dataset and VOiCES dataset. In these experiments, the joint learning of envelope dereverberation and acoustic model yields significant performance improvements over the baseline ASR system based on log-mel spectrogram as well as other past approaches for dereverberation (average relative improvements of 10-24% over the baseline system). A detailed analysis on the choice of hyper-parameters and the cost function involved in envelope dereverberation is also provided.

End-To-End Speech Recognition with Joint Dereverberation of Sub-Band Autoregressive Envelopes

“…Recently, end-to-end models with attention based modeling have also been explored on the REVERB challenge dataset [21,22]. Previously, we had proposed a convolutional neural network model to perform dereverberation of speech [10,11]. In the current work, we extend this prior work for E2E transformer based ASR system.…”

“…Our previous work [10,11] explored the use of dereverberation of sub-band envelopes for hybrid speech recognition systems. The sub-band envelopes are extracted using the autoregressive modeling framework of frequency domain linear prediction [12,13].…”

Research on Throat Speech Signal Detection Based on a Flexible Graphene Piezoresistive Sensor

“…One is to improve the pickup equipment or pickup method, such as using a microphone array instead of a single microphone, that is, multichannel speech signal acquisition. The second is to use some signal processing methods to improve the quality of far-field speech. , In this paper, from the perspective of front-end pickup, a flexible graphene sensor is proposed to detect vocal cord vibration signals to improve speech recognition performance in the far-field environment. Like other physiological signals, such as breathing signals and pulse signals, vocal fold vibration signals are relatively weak signals.…”

“…The second is to use some signal processing methods to improve the quality of farfield speech. 13,14 In this paper, from the perspective of frontend pickup, a flexible graphene sensor is proposed to detect short vowels in human pronunciation using graphene-like MXene to detect vocal vibration signals. However, the preparation process of this material is complex and the cost is high, and the stability of the MXene material at room temperature is poor.…”