Page: 1 2 3 4 5 6 7 8 9... 567
81 |
Emotion Intensity and its Control for Emotional Voice Conversion ...
|
|
|
|
BASE
|
|
Show details
|
|
82 |
Automatic Speech recognition for Speech Assessment of Preschool Children ...
|
|
|
|
BASE
|
|
Show details
|
|
83 |
Deep Speech Based End-to-End Automated Speech Recognition (ASR) for Indian-English Accents ...
|
|
|
|
BASE
|
|
Show details
|
|
84 |
KazakhTTS2: Extending the Open-Source Kazakh TTS Corpus With More Data, Speakers, and Topics ...
|
|
|
|
BASE
|
|
Show details
|
|
85 |
Automated speech tools for helping communities process restricted-access corpora for language revival efforts ...
|
|
|
|
BASE
|
|
Show details
|
|
87 |
Separate What You Describe: Language-Queried Audio Source Separation ...
|
|
|
|
BASE
|
|
Show details
|
|
88 |
A Complementary Joint Training Approach Using Unpaired Speech and Text for Low-Resource Automatic Speech Recognition ...
|
|
|
|
BASE
|
|
Show details
|
|
90 |
The First Gospel, the Gospel of the Poor: A New Reconstruction of Q and Resolution of the Synoptic Problem based on Marcion's Early Luke ...
|
|
|
|
BASE
|
|
Show details
|
|
97 |
Measuring the Impact of Individual Domain Factors in Self-Supervised Pre-Training ...
|
|
|
|
BASE
|
|
Show details
|
|
98 |
Low-dimensional representation of infant and adult vocalization acoustics ...
|
|
|
|
BASE
|
|
Show details
|
|
99 |
Dual-Decoder Transformer For end-to-end Mandarin Chinese Speech Recognition with Pinyin and Character ...
|
|
|
|
BASE
|
|
Show details
|
|
100 |
Similarity and Content-based Phonetic Self Attention for Speech Recognition ...
|
|
|
|
Abstract:
Transformer-based speech recognition models have achieved great success due to the self-attention (SA) mechanism that utilizes every frame in the feature extraction process. Especially, SA heads in lower layers capture various phonetic characteristics by the query-key dot product, which is designed to compute the pairwise relationship between frames. In this paper, we propose a variant of SA to extract more representative phonetic features. The proposed phonetic self-attention (phSA) is composed of two different types of phonetic attention; one is similarity-based and the other is content-based. In short, similarity-based attention utilizes the correlation between frames while content-based attention only considers each frame without being affected by others. We identify which parts of the original dot product are related to two different attention patterns and improve each part by simple modifications. Our experiments on phoneme classification and speech recognition show that replacing SA with phSA for ... : Submitted to INTERSPEECH 2022 ...
|
|
Keyword:
Audio and Speech Processing eess.AS; Computation and Language cs.CL; FOS Computer and information sciences; FOS Electrical engineering, electronic engineering, information engineering
|
|
URL: https://dx.doi.org/10.48550/arxiv.2203.10252 https://arxiv.org/abs/2203.10252
|
|
BASE
|
|
Hide details
|
|
Page: 1 2 3 4 5 6 7 8 9... 567
|
|