-
Multi-ACCDOA: Localizing and Detecting Overlapping Sounds from the Same Class with Auxiliary Duplicating Permutation Invariant Training
Sound event localization and detection (SELD) involves identifying the d...
10/14/2021 ∙ by Kazuki Shimada, et al. ∙
0
∙
share
read it
-
Spatial Data Augmentation with Simulated Room Impulse Responses for Sound Event Localization and Detection
Recording and annotating real sound events for a sound event localizatio...
10/13/2021 ∙ by Yuichiro Koyama, et al. ∙
0
∙
share
read it
-
Ensemble of ACCDOA- and EINV2-based Systems with D3Nets and Impulse Response Simulation for Sound Event Localization and Detection
This report describes our systems submitted to the DCASE2021 challenge t...
06/21/2021 ∙ by Kazuki Shimada, et al. ∙
0
∙
share
read it
-
Data Augmentation Methods for End-to-end Speech Recognition on Distant-Talk Scenarios
Although end-to-end automatic speech recognition (E2E ASR) has achieved ...
06/07/2021 ∙ by Emiru Tsunoo, et al. ∙
0
∙
share
read it
-
Gaussian Kernelized Self-Attention for Long Sequence Data and Its Application to CTC-based Speech Recognition
Self-attention (SA) based models have recently achieved significant perf...
02/18/2021 ∙ by Yosuke Kashiwagi, et al. ∙
0
∙
share
read it
-
Streaming Transformer ASR with Blockwise Synchronous Inference
The Transformer self-attention network has recently shown promising perf...
06/25/2020 ∙ by Emiru Tsunoo, et al. ∙
0
∙
share
read it
-
Towards Online End-to-end Transformer Automatic Speech Recognition
The Transformer self-attention network has recently shown promising perf...
10/25/2019 ∙ by Emiru Tsunoo, et al. ∙
0
∙
share
read it
-
Transformer ASR with Contextual Block Processing
The Transformer self-attention network has recently shown promising perf...
10/16/2019 ∙ by Emiru Tsunoo, et al. ∙
0
∙
share
read it
-
End-to-end Adaptation with Backpropagation through WFST for On-device Speech Recognition System
An on-device DNN-HMM speech recognition system efficiently works with a ...
05/17/2019 ∙ by Emiru Tsunoo, et al. ∙
0
∙
share
read it
