Receptive Field Analysis of Temporal Convolutional Networks for Monaural Speech Dereverberation

13 Apr 2022  ยท  William Ravenscroft, Stefan Goetze, Thomas Hain ยท

Speech dereverberation is often an important requirement in robust speech processing tasks. Supervised deep learning (DL) models give state-of-the-art performance for single-channel speech dereverberation. Temporal convolutional networks (TCNs) are commonly used for sequence modelling in speech enhancement tasks. A feature of TCNs is that they have a receptive field (RF) dependent on the specific model configuration which determines the number of input frames that can be observed to produce an individual output frame. It has been shown that TCNs are capable of performing dereverberation of simulated speech data, however a thorough analysis, especially with focus on the RF is yet lacking in the literature. This paper analyses dereverberation performance depending on the model size and the RF of TCNs. Experiments using the WHAMR corpus which is extended to include room impulse responses (RIRs) with larger T60 values demonstrate that a larger RF can have significant improvement in performance when training smaller TCN models. It is also demonstrated that TCNs benefit from a wider RF when dereverberating RIRs with larger RT60 values.

PDF Abstract

Datasets


Introduced in the Paper:

WHAMR_ext

Used in the Paper:

WHAM! WHAMR!

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Speech Dereverberation WHAMR! Conv-TasNet DAE SI-SDR 12.03 # 2
PESQ 3.46 # 2
SI-SDRi 7.63 # 1
ESTOI 93 # 2
SRMR 8.7 # 2
Speech Dereverberation WHAMR_ext Conv-TasNet DAE SI-SDR 7.07 # 1
SI-SDRi 10.81 # 1
PESQ 2.46 # 1
ESTOI 81 # 1
SRMR 9.18 # 1

Methods


No methods listed for this paper. Add relevant methods here