paper_id
stringlengths
15
35
title
stringlengths
26
182
authors
listlengths
1
25
isca_url
stringlengths
66
86
pdf_url
stringlengths
65
85
doi
stringlengths
27
30
pages
stringlengths
3
9
bibtex
large_stringlengths
294
850
abstract
large_stringlengths
247
1.59k
arxiv_id
stringlengths
10
10
arxiv_id_source
stringclasses
2 values
allauzen21_interspeech
A Hybrid Seq-2-Seq ASR Design for On-Device and Server Applications
[ "Cyril Allauzen", "Ehsan Variani", "Michael Riley", "David Rybach", "Hao Zhang" ]
https://www.isca-archive.org/interspeech_2021/allauzen21_interspeech.html
https://www.isca-archive.org/interspeech_2021/allauzen21_interspeech.pdf
10.21437/Interspeech.2021-658
4044-4048
@inproceedings{allauzen21_interspeech, title = {{A Hybrid Seq-2-Seq ASR Design for On-Device and Server Applications}}, author = {Cyril Allauzen and Ehsan Variani and Michael Riley and David Rybach and Hao Zhang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4044--4048}, doi ...
This paper proposes and evaluates alternative speech recognition design strategies using the hybrid autoregressive transducer (HAT) model. The different strategies are designed with special attention to the choice of modeling units and to the integration of different types of external language models during first-pass ...
null
null
inaguma21b_interspeech
VAD-Free Streaming Hybrid CTC/Attention ASR for Unsegmented Recording
[ "Hirofumi Inaguma", "Tatsuya Kawahara" ]
https://www.isca-archive.org/interspeech_2021/inaguma21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/inaguma21b_interspeech.pdf
10.21437/Interspeech.2021-1107
4049-4053
@inproceedings{inaguma21b_interspeech, title = {{VAD-Free Streaming Hybrid CTC/Attention ASR for Unsegmented Recording}}, author = {Hirofumi Inaguma and Tatsuya Kawahara}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4049--4053}, doi = {10.21437/Interspeech.2021-1107}, i...
In this work, we propose novel decoding algorithms to enable streaming automatic speech recognition (ASR) on unsegmented long-form recordings without voice activity detection (VAD), based on monotonic chunkwise attention (MoChA) with an auxiliary connectionist temporal classification (CTC) objective. We propose a block...
2107.07509
title_snapshot
yao21_interspeech
WeNet: Production Oriented Streaming and Non-Streaming End-to-End Speech Recognition Toolkit
[ "Zhuoyuan Yao", "Di Wu", "Xiong Wang", "Binbin Zhang", "Fan Yu", "Chao Yang", "Zhendong Peng", "Xiaoyu Chen", "Lei Xie", "Xin Lei" ]
https://www.isca-archive.org/interspeech_2021/yao21_interspeech.html
https://www.isca-archive.org/interspeech_2021/yao21_interspeech.pdf
10.21437/Interspeech.2021-1983
4054-4058
@inproceedings{yao21_interspeech, title = {{WeNet: Production Oriented Streaming and Non-Streaming End-to-End Speech Recognition Toolkit}}, author = {Zhuoyuan Yao and Di Wu and Xiong Wang and Binbin Zhang and Fan Yu and Chao Yang and Zhendong Peng and Xiaoyu Chen and Lei Xie and Xin Lei}, year = {2021...
In this paper, we propose an open source speech recognition toolkit called WeNet, in which a new two-pass approach named U2 is implemented to unify streaming and non-streaming end-to-end (E2E) speech recognition in a single model. The main motivation of WeNet is to close the gap between the research and deployment of E...
2102.01547
title_snapshot
tanaka21b_interspeech
Cross-Modal Transformer-Based Neural Correction Models for Automatic Speech Recognition
[ "Tomohiro Tanaka", "Ryo Masumura", "Mana Ihori", "Akihiko Takashima", "Takafumi Moriya", "Takanori Ashihara", "Shota Orihashi", "Naoki Makishima" ]
https://www.isca-archive.org/interspeech_2021/tanaka21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/tanaka21b_interspeech.pdf
10.21437/Interspeech.2021-1992
4059-4063
@inproceedings{tanaka21b_interspeech, title = {{Cross-Modal Transformer-Based Neural Correction Models for Automatic Speech Recognition}}, author = {Tomohiro Tanaka and Ryo Masumura and Mana Ihori and Akihiko Takashima and Takafumi Moriya and Takanori Ashihara and Shota Orihashi and Naoki Makishima}, year ...
We propose a cross-modal transformer-based neural correction models that refines the output of an automatic speech recognition (ASR) system so as to exclude ASR errors. Generally, neural correction models are composed of encoder-decoder networks, which can directly model sequence-to-sequence mapping problems. The most ...
2107.01569
title_snapshot
lee21f_interspeech
Deep Neural Network Calibration for E2E Speech Recognition System
[ "Mun-Hak Lee", "Joon-Hyuk Chang" ]
https://www.isca-archive.org/interspeech_2021/lee21f_interspeech.html
https://www.isca-archive.org/interspeech_2021/lee21f_interspeech.pdf
10.21437/Interspeech.2021-176
4064-4068
@inproceedings{lee21f_interspeech, title = {{Deep Neural Network Calibration for E2E Speech Recognition System}}, author = {Mun-Hak Lee and Joon-Hyuk Chang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4064--4068}, doi = {10.21437/Interspeech.2021-176}, issn = {295...
Cross-entropy loss, which is commonly used in deep-neural-network-based (DNN) classification model training, induces models to assign a high probability value to one class. Networks trained in this fashion tend to be overconfident, which causes a problem in the decoding process of the speech recognition system, as it u...
null
null
li21m_interspeech
Residual Energy-Based Models for End-to-End Speech Recognition
[ "Qiujia Li", "Yu Zhang", "Bo Li", "Liangliang Cao", "Philip C. Woodland" ]
https://www.isca-archive.org/interspeech_2021/li21m_interspeech.html
https://www.isca-archive.org/interspeech_2021/li21m_interspeech.pdf
10.21437/Interspeech.2021-690
4069-4073
@inproceedings{li21m_interspeech, title = {{Residual Energy-Based Models for End-to-End Speech Recognition}}, author = {Qiujia Li and Yu Zhang and Bo Li and Liangliang Cao and Philip C. Woodland}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4069--4073}, doi = {10.21437/In...
End-to-end models with auto-regressive decoders have shown impressive results for automatic speech recognition (ASR). These models formulate the sequence-level probability as a product of the conditional probabilities of all individual tokens given their histories. However, the performance of locally normalised models ...
2103.14152
title_snapshot
qiu21b_interspeech
Multi-Task Learning for End-to-End ASR Word and Utterance Confidence with Deletion Prediction
[ "David Qiu", "Yanzhang He", "Qiujia Li", "Yu Zhang", "Liangliang Cao", "Ian McGraw" ]
https://www.isca-archive.org/interspeech_2021/qiu21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/qiu21b_interspeech.pdf
10.21437/Interspeech.2021-1207
4074-4078
@inproceedings{qiu21b_interspeech, title = {{Multi-Task Learning for End-to-End ASR Word and Utterance Confidence with Deletion Prediction}}, author = {David Qiu and Yanzhang He and Qiujia Li and Yu Zhang and Liangliang Cao and Ian McGraw}, year = {2021}, booktitle = {{Interspeech 2021}}, pages ...
Confidence scores are very useful for downstream applications of automatic speech recognition (ASR) systems. Recent works have proposed using neural networks to learn word or utterance confidence scores for end-to-end ASR. In those studies, word confidence by itself does not model deletions, and utterance confidence do...
2104.12870
title_snapshot
ollerenshaw21_interspeech
Insights on Neural Representations for End-to-End Speech Recognition
[ "Anna Ollerenshaw", "Md. Asif Jalal", "Thomas Hain" ]
https://www.isca-archive.org/interspeech_2021/ollerenshaw21_interspeech.html
https://www.isca-archive.org/interspeech_2021/ollerenshaw21_interspeech.pdf
10.21437/Interspeech.2021-1516
4079-4083
@inproceedings{ollerenshaw21_interspeech, title = {{Insights on Neural Representations for End-to-End Speech Recognition}}, author = {Anna Ollerenshaw and Md. Asif Jalal and Thomas Hain}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4079--4083}, doi = {10.21437/Interspeech...
End-to-end automatic speech recognition (ASR) models aim to learn a generalised speech representation. However, there are limited tools available to understand the internal functions and the effect of hierarchical dependencies within the model architecture. It is crucial to understand the correlations between the layer...
2205.09456
title_snapshot
afshan21_interspeech
Sequence-Level Confidence Classifier for ASR Utterance Accuracy and Application to Acoustic Models
[ "Amber Afshan", "Kshitiz Kumar", "Jian Wu" ]
https://www.isca-archive.org/interspeech_2021/afshan21_interspeech.html
https://www.isca-archive.org/interspeech_2021/afshan21_interspeech.pdf
10.21437/Interspeech.2021-1666
4084-4088
@inproceedings{afshan21_interspeech, title = {{Sequence-Level Confidence Classifier for ASR Utterance Accuracy and Application to Acoustic Models}}, author = {Amber Afshan and Kshitiz Kumar and Jian Wu}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4084--4088}, doi = {10.2...
Scores from traditional confidence classifiers (CCs) in automatic speech recognition (ASR) systems lack universal interpretation and vary with updates to the underlying confidence or acoustic models (AMs). In this work, we build interpretable confidence scores with an objective to closely align with ASR accuracy. We pr...
2107.00099
title_snapshot
tjandra21_interspeech
Unsupervised Learning of Disentangled Speech Content and Style Representation
[ "Andros Tjandra", "Ruoming Pang", "Yu Zhang", "Shigeki Karita" ]
https://www.isca-archive.org/interspeech_2021/tjandra21_interspeech.html
https://www.isca-archive.org/interspeech_2021/tjandra21_interspeech.pdf
10.21437/Interspeech.2021-1936
4089-4093
@inproceedings{tjandra21_interspeech, title = {{Unsupervised Learning of Disentangled Speech Content and Style Representation}}, author = {Andros Tjandra and Ruoming Pang and Yu Zhang and Shigeki Karita}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4089--4093}, doi = {10....
Speech is influenced by a number of underlying factors, which can be broadly categorized into linguistic contents and speaking styles. However, collecting the labeled data that annotates both content and style is an expensive and time-consuming task. Here, we present an approach for unsupervised learning of speech repr...
2010.12973
title_snapshot
choi21_interspeech
Label Embedding for Chinese Grapheme-to-Phoneme Conversion
[ "Eunbi Choi", "Hwa-Yeon Kim", "Jong-Hwan Kim", "Jae-Min Kim" ]
https://www.isca-archive.org/interspeech_2021/choi21_interspeech.html
https://www.isca-archive.org/interspeech_2021/choi21_interspeech.pdf
10.21437/Interspeech.2021-885
4094-4098
@inproceedings{choi21_interspeech, title = {{Label Embedding for Chinese Grapheme-to-Phoneme Conversion}}, author = {Eunbi Choi and Hwa-Yeon Kim and Jong-Hwan Kim and Jae-Min Kim}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4094--4098}, doi = {10.21437/Interspeech.2021-8...
Chinese grapheme-to-phoneme (G2P) conversion plays a significant role in text-to-speech systems by generating pronunciations corresponding to Chinese input characters. The main challenge in Chinese G2P conversion is polyphone disambiguation, which requires selecting the appropriate pronunciation among several candidate...
null
null
zhang21aa_interspeech
PDF: Polyphone Disambiguation in Chinese by Using FLAT
[ "Haiteng Zhang" ]
https://www.isca-archive.org/interspeech_2021/zhang21aa_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhang21aa_interspeech.pdf
10.21437/Interspeech.2021-1087
4099-4103
@inproceedings{zhang21aa_interspeech, title = {{PDF: Polyphone Disambiguation in Chinese by Using FLAT}}, author = {Haiteng Zhang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4099--4103}, doi = {10.21437/Interspeech.2021-1087}, issn = {2958-1796}, }
Polyphone disambiguation is an essential procedure in the front-end module of the Chinese text-to-speech (TTS) system. It serves to predict the pronunciation of the input polyphonic character. In the Chinese TTS system, a well-designed pronunciation dictionary plays a crucial role in supplying pinyin to words. However,...
null
null
li21n_interspeech
Improving Polyphone Disambiguation for Mandarin Chinese by Combining Mix-Pooling Strategy and Window-Based Attention
[ "Junjie Li", "Zhiyu Zhang", "Minchuan Chen", "Jun Ma", "Shaojun Wang", "Jing Xiao" ]
https://www.isca-archive.org/interspeech_2021/li21n_interspeech.html
https://www.isca-archive.org/interspeech_2021/li21n_interspeech.pdf
10.21437/Interspeech.2021-1232
4104-4108
@inproceedings{li21n_interspeech, title = {{Improving Polyphone Disambiguation for Mandarin Chinese by Combining Mix-Pooling Strategy and Window-Based Attention}}, author = {Junjie Li and Zhiyu Zhang and Minchuan Chen and Jun Ma and Shaojun Wang and Jing Xiao}, year = {2021}, booktitle = {{Interspee...
In this paper, we propose a novel system based on word-level features and window-based attention for polyphone disambiguation, which is a fundamental task for Grapheme-to-phoneme (G2P) conversion of Mandarin Chinese. The framework aims to combine a pre-trained language model with explicit word-level information in orde...
null
null
shi21d_interspeech
Polyphone Disambiguation in Mandarin Chinese with Semi-Supervised Learning
[ "Yi Shi", "Congyi Wang", "Yu Chen", "Bin Wang" ]
https://www.isca-archive.org/interspeech_2021/shi21d_interspeech.html
https://www.isca-archive.org/interspeech_2021/shi21d_interspeech.pdf
10.21437/Interspeech.2021-502
4109-4113
@inproceedings{shi21d_interspeech, title = {{Polyphone Disambiguation in Mandarin Chinese with Semi-Supervised Learning}}, author = {Yi Shi and Congyi Wang and Yu Chen and Bin Wang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4109--4113}, doi = {10.21437/Interspeech.2021...
The majority of Chinese characters are monophonic, while a special group of characters, called polyphonic characters, have multiple pronunciations. As a prerequisite of performing speech-related generative tasks, the correct pronunciation must be identified among several candidates. This process is called Polyphone Dis...
2102.00621
title_snapshot
chen21s_interspeech
A Neural-Network-Based Approach to Identifying Speakers in Novels
[ "Yue Chen", "Zhen-Hua Ling", "Qing-Feng Liu" ]
https://www.isca-archive.org/interspeech_2021/chen21s_interspeech.html
https://www.isca-archive.org/interspeech_2021/chen21s_interspeech.pdf
10.21437/Interspeech.2021-609
4114-4118
@inproceedings{chen21s_interspeech, title = {{A Neural-Network-Based Approach to Identifying Speakers in Novels}}, author = {Yue Chen and Zhen-Hua Ling and Qing-Feng Liu}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4114--4118}, doi = {10.21437/Interspeech.2021-609}, is...
Identifying speakers in novels aims at determining who says a quote in a given context by text analysis. This task is important for speech synthesis systems to assign appropriate voices to the quotes when producing audiobooks. However, existing approaches stick with manual features and traditional machine learning clas...
null
null
zhou21f_interspeech
UnitNet-Based Hybrid Speech Synthesis
[ "Xiao Zhou", "Zhen-Hua Ling", "Li-Rong Dai" ]
https://www.isca-archive.org/interspeech_2021/zhou21f_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhou21f_interspeech.pdf
10.21437/Interspeech.2021-1092
4119-4123
@inproceedings{zhou21f_interspeech, title = {{UnitNet-Based Hybrid Speech Synthesis}}, author = {Xiao Zhou and Zhen-Hua Ling and Li-Rong Dai}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4119--4123}, doi = {10.21437/Interspeech.2021-1092}, issn = {2958-1796}, }
This paper presents a hybrid speech synthesis method based on UnitNet, a unified sequence-to-sequence (Seq2Seq) acoustic model for both statistical parametric speech synthesis (SPSS) and concatenative speech synthesis (CSS). This method combines CSS and SPSS approaches to synthesize different segments in an utterance. ...
null
null
novitasari21_interspeech
Dynamically Adaptive Machine Speech Chain Inference for TTS in Noisy Environment: Listen and Speak Louder
[ "Sashi Novitasari", "Sakriani Sakti", "Satoshi Nakamura" ]
https://www.isca-archive.org/interspeech_2021/novitasari21_interspeech.html
https://www.isca-archive.org/interspeech_2021/novitasari21_interspeech.pdf
10.21437/Interspeech.2021-946
4124-4128
@inproceedings{novitasari21_interspeech, title = {{Dynamically Adaptive Machine Speech Chain Inference for TTS in Noisy Environment: Listen and Speak Louder}}, author = {Sashi Novitasari and Sakriani Sakti and Satoshi Nakamura}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4124--4...
Although machine speech chains were originally proposed to mimic a closed-loop human speech chain mechanism with auditory feedback, the existing machine speech chains are only utilized as a semi-supervised learning method that allows automatic speech recognition (ASR) and text-to-speech synthesis systems (TTS) to suppo...
null
null
zhang21ba_interspeech
LinearSpeech: Parallel Text-to-Speech with Linear Complexity
[ "Haozhe Zhang", "Zhihua Huang", "Zengqiang Shang", "Pengyuan Zhang", "Yonghong Yan" ]
https://www.isca-archive.org/interspeech_2021/zhang21ba_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhang21ba_interspeech.pdf
10.21437/Interspeech.2021-1192
4129-4133
@inproceedings{zhang21ba_interspeech, title = {{LinearSpeech: Parallel Text-to-Speech with Linear Complexity}}, author = {Haozhe Zhang and Zhihua Huang and Zengqiang Shang and Pengyuan Zhang and Yonghong Yan}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4129--4133}, doi =...
Non-autoregressive text to speech models such as FastSpeech can synthesize speech significantly faster than previous autoregressive models with comparable quality. However, the memory and time complexity O(N 2 ) of self-attention hinders FastSpeech from generating long sequences, where N is the length of mel-spectrogra...
null
null
mansbach21_interspeech
An Agent for Competing with Humans in a Deceptive Game Based on Vocal Cues
[ "Noa Mansbach", "Evgeny Hershkovitch Neiterman", "Amos Azaria" ]
https://www.isca-archive.org/interspeech_2021/mansbach21_interspeech.html
https://www.isca-archive.org/interspeech_2021/mansbach21_interspeech.pdf
10.21437/Interspeech.2021-83
4134-4138
@inproceedings{mansbach21_interspeech, title = {{An Agent for Competing with Humans in a Deceptive Game Based on Vocal Cues}}, author = {Noa Mansbach and Evgeny Hershkovitch Neiterman and Amos Azaria}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4134--4138}, doi = {10.214...
In this work we present the development of an autonomous agent capable of competing with humans in a deception-based game. The agent predicts whether a given statement is true or false based on vocal cues. To this end, we develop a game for collecting a large scale and high quality labeled sound data-set in a controlle...
null
null
fakhry21_interspeech
A Multi-Branch Deep Learning Network for Automated Detection of COVID-19
[ "Ahmed Fakhry", "Xinyi Jiang", "Jaclyn Xiao", "Gunvant Chaudhari", "Asriel Han" ]
https://www.isca-archive.org/interspeech_2021/fakhry21_interspeech.html
https://www.isca-archive.org/interspeech_2021/fakhry21_interspeech.pdf
10.21437/Interspeech.2021-378
4139-4143
@inproceedings{fakhry21_interspeech, title = {{A Multi-Branch Deep Learning Network for Automated Detection of COVID-19}}, author = {Ahmed Fakhry and Xinyi Jiang and Jaclyn Xiao and Gunvant Chaudhari and Asriel Han}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4139--4143}, doi ...
Fast and affordable solutions for COVID-19 testing are necessary to contain the spread of the global pandemic and help relieve the burden on medical facilities. Currently, limited testing locations and expensive equipment pose difficulties for individuals seeking testing, especially in low-resource settings. Researcher...
null
null
ma21d_interspeech
RW-Resnet: A Novel Speech Anti-Spoofing Model Using Raw Waveform
[ "Youxuan Ma", "Zongze Ren", "Shugong Xu" ]
https://www.isca-archive.org/interspeech_2021/ma21d_interspeech.html
https://www.isca-archive.org/interspeech_2021/ma21d_interspeech.pdf
10.21437/Interspeech.2021-438
4144-4148
@inproceedings{ma21d_interspeech, title = {{RW-Resnet: A Novel Speech Anti-Spoofing Model Using Raw Waveform}}, author = {Youxuan Ma and Zongze Ren and Shugong Xu}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4144--4148}, doi = {10.21437/Interspeech.2021-438}, issn ...
In recent years, synthetic speech generated by advanced text-to-speech (TTS) and voice conversion (VC) systems has caused great harms to automatic speaker verification (ASV) systems, urging us to design a synthetic speech detection system to protect ASV systems. In this paper, we propose a new speech anti-spoofing mode...
2108.05684
title_snapshot
dhamyal21_interspeech
Fake Audio Detection in Resource-Constrained Settings Using Microfeatures
[ "Hira Dhamyal", "Ayesha Ali", "Ihsan Ayyub Qazi", "Agha Ali Raza" ]
https://www.isca-archive.org/interspeech_2021/dhamyal21_interspeech.html
https://www.isca-archive.org/interspeech_2021/dhamyal21_interspeech.pdf
10.21437/Interspeech.2021-524
4149-4153
@inproceedings{dhamyal21_interspeech, title = {{Fake Audio Detection in Resource-Constrained Settings Using Microfeatures}}, author = {Hira Dhamyal and Ayesha Ali and Ihsan Ayyub Qazi and Agha Ali Raza}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4149--4153}, doi = {10.2...
Fake audio generation has undergone remarkable improvement with the advancement in deep neural network models. This has made it increasingly important to develop lightweight yet robust mechanisms for detecting fake audios, especially for resource-constrained settings such as on edge devices and embedded controllers as ...
null
null
yan21c_interspeech
Coughing-Based Recognition of Covid-19 with Spatial Attentive ConvLSTM Recurrent Neural Networks
[ "Tianhao Yan", "Hao Meng", "Emilia Parada-Cabaleiro", "Shuo Liu", "Meishu Song", "Björn W. Schuller" ]
https://www.isca-archive.org/interspeech_2021/yan21c_interspeech.html
https://www.isca-archive.org/interspeech_2021/yan21c_interspeech.pdf
10.21437/Interspeech.2021-630
4154-4158
@inproceedings{yan21c_interspeech, title = {{Coughing-Based Recognition of Covid-19 with Spatial Attentive ConvLSTM Recurrent Neural Networks}}, author = {Tianhao Yan and Hao Meng and Emilia Parada-Cabaleiro and Shuo Liu and Meishu Song and Björn W. Schuller}, year = {2021}, booktitle = {{Interspeec...
The rapid emergence of COVID-19 has become a major public health threat around the world. Although early detection is crucial to reduce its spread, the existing diagnostic methods are still insufficient in bringing the pandemic under control. Thus, more sophisticated systems, able to easily identify the infection from ...
null
null
paul21b_interspeech
Knowledge Distillation for Singing Voice Detection
[ "Soumava Paul", "Gurunath Reddy M", "K. Sreenivasa Rao", "Partha Pratim Das" ]
https://www.isca-archive.org/interspeech_2021/paul21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/paul21b_interspeech.pdf
10.21437/Interspeech.2021-636
4159-4163
@inproceedings{paul21b_interspeech, title = {{Knowledge Distillation for Singing Voice Detection}}, author = {Soumava Paul and Gurunath Reddy M and K. Sreenivasa Rao and Partha Pratim Das}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4159--4163}, doi = {10.21437/Interspee...
Singing Voice Detection (SVD) has been an active area of research in music information retrieval (MIR). Currently, two deep neural network-based methods, one based on CNN and the other on RNN, exist in literature that learn optimized features for the voice detection (VD) task and achieve state-of-the-art performance on...
2011.04297
title_snapshot
takeda21_interspeech
Age Estimation with Speech-Age Model for Heterogeneous Speech Datasets
[ "Ryu Takeda", "Kazunori Komatani" ]
https://www.isca-archive.org/interspeech_2021/takeda21_interspeech.html
https://www.isca-archive.org/interspeech_2021/takeda21_interspeech.pdf
10.21437/Interspeech.2021-861
4164-4168
@inproceedings{takeda21_interspeech, title = {{Age Estimation with Speech-Age Model for Heterogeneous Speech Datasets}}, author = {Ryu Takeda and Kazunori Komatani}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4164--4168}, doi = {10.21437/Interspeech.2021-861}, issn ...
This paper describes an age estimation method from speech signals for heterogeneous datasets. Although previous studies in the speech field evaluate age prediction models with held-out testing data within the same dataset recorded in a consistent setting, such evaluation does not measure real performance. The difficult...
null
null
teh21_interspeech
Open-Set Audio Classification with Limited Training Resources Based on Augmentation Enhanced Variational Auto-Encoder GAN with Detection-Classification Joint Training
[ "Kah Kuan Teh", "Huy Dat Tran" ]
https://www.isca-archive.org/interspeech_2021/teh21_interspeech.html
https://www.isca-archive.org/interspeech_2021/teh21_interspeech.pdf
10.21437/Interspeech.2021-1142
4169-4173
@inproceedings{teh21_interspeech, title = {{Open-Set Audio Classification with Limited Training Resources Based on Augmentation Enhanced Variational Auto-Encoder GAN with Detection-Classification Joint Training}}, author = {Kah Kuan Teh and Huy Dat Tran}, year = {2021}, booktitle = {{Interspeech 202...
In this paper, we propose a novel method to address practical problems when deploying audio classification systems in operations that are the presence of unseen sound classes (open-set) and the limitation of training resources. To solve it, a novel method which embeds variational auto-encoder (VAE), data augmentation a...
null
null
fukumori21_interspeech
Deep Spectral-Cepstral Fusion for Shouted and Normal Speech Classification
[ "Takahiro Fukumori" ]
https://www.isca-archive.org/interspeech_2021/fukumori21_interspeech.html
https://www.isca-archive.org/interspeech_2021/fukumori21_interspeech.pdf
10.21437/Interspeech.2021-1245
4174-4178
@inproceedings{fukumori21_interspeech, title = {{Deep Spectral-Cepstral Fusion for Shouted and Normal Speech Classification}}, author = {Takahiro Fukumori}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4174--4178}, doi = {10.21437/Interspeech.2021-1245}, issn = {295...
Discrimination between shouted and normal speech is crucial in audio surveillance and monitoring. Although deep neural networks are used in recent methods, traditional low-level speech features are applied, such as mel-frequency cepstral coefficients and the mel spectrum. This paper presents a deep spectral-cepstral fu...
null
null
baghel21_interspeech
Automatic Detection of Shouted Speech Segments in Indian News Debates
[ "Shikha Baghel", "Mrinmoy Bhattacharjee", "S.R. Mahadeva Prasanna", "Prithwijit Guha" ]
https://www.isca-archive.org/interspeech_2021/baghel21_interspeech.html
https://www.isca-archive.org/interspeech_2021/baghel21_interspeech.pdf
10.21437/Interspeech.2021-1592
4179-4183
@inproceedings{baghel21_interspeech, title = {{Automatic Detection of Shouted Speech Segments in Indian News Debates}}, author = {Shikha Baghel and Mrinmoy Bhattacharjee and S.R. Mahadeva Prasanna and Prithwijit Guha}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4179--4183}, do...
Shouted speech detection is an essential pre-processing step in conventional speech processing systems such as speech and speaker recognition, speaker diarization, and others. Excitation source plays an important role in shouted speech production. This work explores feature computed from the Integrated Linear Predictio...
null
null
gao21c_interspeech
Generalized Spoofing Detection Inspired from Audio Generation Artifacts
[ "Yang Gao", "Tyler Vuong", "Mahsa Elyasi", "Gaurav Bharaj", "Rita Singh" ]
https://www.isca-archive.org/interspeech_2021/gao21c_interspeech.html
https://www.isca-archive.org/interspeech_2021/gao21c_interspeech.pdf
10.21437/Interspeech.2021-1705
4184-4188
@inproceedings{gao21c_interspeech, title = {{Generalized Spoofing Detection Inspired from Audio Generation Artifacts}}, author = {Yang Gao and Tyler Vuong and Mahsa Elyasi and Gaurav Bharaj and Rita Singh}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4184--4188}, doi = {1...
State-of-the-art methods for audio generation suffer from fingerprint artifacts and repeated inconsistencies across temporal and spectral domains. Such artifacts could be well captured by the frequency domain analysis over the spectrogram. Thus, we propose a novel use of long-range spectro-temporal modulation feature —...
2104.04111
title_snapshot
chen21t_interspeech
Overlapped Speech Detection Based on Spectral and Spatial Feature Fusion
[ "Weiguang Chen", "Van Tung Pham", "Eng Siong Chng", "Xionghu Zhong" ]
https://www.isca-archive.org/interspeech_2021/chen21t_interspeech.html
https://www.isca-archive.org/interspeech_2021/chen21t_interspeech.pdf
10.21437/Interspeech.2021-2138
4189-4193
@inproceedings{chen21t_interspeech, title = {{Overlapped Speech Detection Based on Spectral and Spatial Feature Fusion}}, author = {Weiguang Chen and Van Tung Pham and Eng Siong Chng and Xionghu Zhong}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4189--4193}, doi = {10.21...
Overlapped speech is widely present in conversations and can cause significant performance degradation on speech processing such as diarization, enhancement, and recognition. Detection of overlapped speech, in particular when the speakers are in the far-field, is a challenging task as the overlapped part is usually sho...
null
null
abdullah21_interspeech
Do Acoustic Word Embeddings Capture Phonological Similarity? An Empirical Study
[ "Badr M. Abdullah", "Marius Mosbach", "Iuliia Zaitova", "Bernd Möbius", "Dietrich Klakow" ]
https://www.isca-archive.org/interspeech_2021/abdullah21_interspeech.html
https://www.isca-archive.org/interspeech_2021/abdullah21_interspeech.pdf
10.21437/Interspeech.2021-678
4194-4198
@inproceedings{abdullah21_interspeech, title = {{Do Acoustic Word Embeddings Capture Phonological Similarity? An Empirical Study}}, author = {Badr M. Abdullah and Marius Mosbach and Iuliia Zaitova and Bernd Möbius and Dietrich Klakow}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {...
Several variants of deep neural networks have been successfully employed for building parametric models that project variable-duration spoken word segments onto fixed-size vector representations, or acoustic word embeddings (AWEs). However, it remains unclear to what degree we can rely on the distance in the emerging A...
2106.08686
title_snapshot
gao21d_interspeech
Paraphrase Label Alignment for Voice Application Retrieval in Spoken Language Understanding
[ "Zheng Gao", "Radhika Arava", "Qian Hu", "Xibin Gao", "Thahir Mohamed", "Wei Xiao", "Mohamed AbdelHady" ]
https://www.isca-archive.org/interspeech_2021/gao21d_interspeech.html
https://www.isca-archive.org/interspeech_2021/gao21d_interspeech.pdf
10.21437/Interspeech.2021-97
4199-4203
@inproceedings{gao21d_interspeech, title = {{Paraphrase Label Alignment for Voice Application Retrieval in Spoken Language Understanding}}, author = {Zheng Gao and Radhika Arava and Qian Hu and Xibin Gao and Thahir Mohamed and Wei Xiao and Mohamed AbdelHady}, year = {2021}, booktitle = {{Interspeech...
Spoken language understanding (SLU) smart assistants such as Amazon Alexa host hundreds of thousands of voice applications (skills) to delight end-users and fulfill their utterance requests. Sometimes utterances fail to be claimed by smart assistants due to system problems such as model incapability or routing errors. ...
null
null
rikhye21_interspeech
Personalized Keyphrase Detection Using Speaker and Environment Information
[ "Rajeev Rikhye", "Quan Wang", "Qiao Liang", "Yanzhang He", "Ding Zhao", "Yiteng Huang", "Arun Narayanan", "Ian McGraw" ]
https://www.isca-archive.org/interspeech_2021/rikhye21_interspeech.html
https://www.isca-archive.org/interspeech_2021/rikhye21_interspeech.pdf
10.21437/Interspeech.2021-204
4204-4208
@inproceedings{rikhye21_interspeech, title = {{Personalized Keyphrase Detection Using Speaker and Environment Information}}, author = {Rajeev Rikhye and Quan Wang and Qiao Liang and Yanzhang He and Ding Zhao and Yiteng Huang and Arun Narayanan and Ian McGraw}, year = {2021}, booktitle = {{Interspeec...
In this paper, we introduce a streaming keyphrase detection system that can be easily customized to accurately detect any phrase composed of words from a large vocabulary. The system is implemented with an end-to-end trained automatic speech recognition (ASR) model and a text-independent speaker verification model. To ...
2104.13970
title_snapshot
garg21_interspeech
Streaming Transformer for Hardware Efficient Voice Trigger Detection and False Trigger Mitigation
[ "Vineet Garg", "Wonil Chang", "Siddharth Sigtia", "Saurabh Adya", "Pramod Simha", "Pranay Dighe", "Chandra Dhir" ]
https://www.isca-archive.org/interspeech_2021/garg21_interspeech.html
https://www.isca-archive.org/interspeech_2021/garg21_interspeech.pdf
10.21437/Interspeech.2021-1428
4209-4213
@inproceedings{garg21_interspeech, title = {{Streaming Transformer for Hardware Efficient Voice Trigger Detection and False Trigger Mitigation}}, author = {Vineet Garg and Wonil Chang and Siddharth Sigtia and Saurabh Adya and Pramod Simha and Pranay Dighe and Chandra Dhir}, year = {2021}, booktitle ...
We present a unified and hardware efficient architecture for two stage voice trigger detection (VTD) and false trigger mitigation (FTM) tasks. Two stage VTD systems of voice assistants can get falsely activated to audio segments acoustically similar to the trigger phrase of interest. FTM systems cancel such activations...
2105.06598
title_snapshot
mazumder21_interspeech
Few-Shot Keyword Spotting in Any Language
[ "Mark Mazumder", "Colby Banbury", "Josh Meyer", "Pete Warden", "Vijay Janapa Reddi" ]
https://www.isca-archive.org/interspeech_2021/mazumder21_interspeech.html
https://www.isca-archive.org/interspeech_2021/mazumder21_interspeech.pdf
10.21437/Interspeech.2021-1966
4214-4218
@inproceedings{mazumder21_interspeech, title = {{Few-Shot Keyword Spotting in Any Language}}, author = {Mark Mazumder and Colby Banbury and Josh Meyer and Pete Warden and Vijay Janapa Reddi}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4214--4218}, doi = {10.21437/Intersp...
We introduce a few-shot transfer learning method for keyword spotting in any language. Leveraging open speech corpora in nine languages, we automate the extraction of a large multilingual keyword bank and use it to train an embedding model. With just five training examples, we fine-tune the embedding model for keyword ...
2104.01454
title_snapshot
wang21da_interspeech
Text Anchor Based Metric Learning for Small-Footprint Keyword Spotting
[ "Li Wang", "Rongzhi Gu", "Nuo Chen", "Yuexian Zou" ]
https://www.isca-archive.org/interspeech_2021/wang21da_interspeech.html
https://www.isca-archive.org/interspeech_2021/wang21da_interspeech.pdf
10.21437/Interspeech.2021-136
4219-4223
@inproceedings{wang21da_interspeech, title = {{Text Anchor Based Metric Learning for Small-Footprint Keyword Spotting}}, author = {Li Wang and Rongzhi Gu and Nuo Chen and Yuexian Zou}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4219--4223}, doi = {10.21437/Interspeech.20...
Keyword Spotting (KWS) remains challenging to achieve the trade-off between small footprint and high accuracy. Recently proposed metric learning approaches improved the generalizability of models for the KWS task, and 1D-CNN based KWS models have achieved the state-of-the-arts (SOTA) in terms of model size. However, fo...
2108.05516
title_snapshot
chen21u_interspeech
A Meta-Learning Approach for User-Defined Spoken Term Classification with Varying Classes and Examples
[ "Yangbin Chen", "Tom Ko", "Jianping Wang" ]
https://www.isca-archive.org/interspeech_2021/chen21u_interspeech.html
https://www.isca-archive.org/interspeech_2021/chen21u_interspeech.pdf
10.21437/Interspeech.2021-147
4224-4228
@inproceedings{chen21u_interspeech, title = {{A Meta-Learning Approach for User-Defined Spoken Term Classification with Varying Classes and Examples}}, author = {Yangbin Chen and Tom Ko and Jianping Wang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4224--4228}, doi = {10...
Recently we formulated a user-defined spoken term classification task as a few-shot learning task and tackled the task using Model-Agnostic Meta-Learning (MAML) algorithm. Our results show that the meta-learning approach performs much better than conventional supervised learning and transfer learning in the task, espec...
null
null
lee21g_interspeech
Auxiliary Sequence Labeling Tasks for Disfluency Detection
[ "Dongyub Lee", "Byeongil Ko", "Myeong Cheol Shin", "Taesun Whang", "Daniel Lee", "Eunhwa Kim", "Eunggyun Kim", "Jaechoon Jo" ]
https://www.isca-archive.org/interspeech_2021/lee21g_interspeech.html
https://www.isca-archive.org/interspeech_2021/lee21g_interspeech.pdf
10.21437/Interspeech.2021-400
4229-4233
@inproceedings{lee21g_interspeech, title = {{Auxiliary Sequence Labeling Tasks for Disfluency Detection}}, author = {Dongyub Lee and Byeongil Ko and Myeong Cheol Shin and Taesun Whang and Daniel Lee and Eunhwa Kim and Eunggyun Kim and Jaechoon Jo}, year = {2021}, booktitle = {{Interspeech 2021}}, ...
Detecting disfluencies in spontaneous speech is an important preprocessing step in natural language processing and speech recognition applications. Existing works for disfluency detection have focused on designing a single objective only for disfluency detection, while auxiliary objectives utilizing linguistic informat...
2011.04512
title_snapshot
zhou21g_interspeech
Energy-Friendly Keyword Spotting System Using Add-Based Convolution
[ "Hang Zhou", "Wenchao Hu", "Yu Ting Yeung", "Xiao Chen" ]
https://www.isca-archive.org/interspeech_2021/zhou21g_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhou21g_interspeech.pdf
10.21437/Interspeech.2021-458
4234-4238
@inproceedings{zhou21g_interspeech, title = {{Energy-Friendly Keyword Spotting System Using Add-Based Convolution}}, author = {Hang Zhou and Wenchao Hu and Yu Ting Yeung and Xiao Chen}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4234--4238}, doi = {10.21437/Interspeech.2...
Wake-up keyword of a keyword spotting (KWS) system represents brand name of a smart device. Performance of KWS is also crucial for modern speech based human-device interaction. An on-device KWS with both high accuracy and low power consumption is desired. We propose a KWS with add-based convolution layers, namely Add T...
null
null
jia21b_interspeech
The 2020 Personalized Voice Trigger Challenge: Open Datasets, Evaluation Metrics, Baseline System and Results
[ "Yan Jia", "Xingming Wang", "Xiaoyi Qin", "Yinping Zhang", "Xuyang Wang", "Junjie Wang", "Dong Zhang", "Ming Li" ]
https://www.isca-archive.org/interspeech_2021/jia21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/jia21b_interspeech.pdf
10.21437/Interspeech.2021-602
4239-4243
@inproceedings{jia21b_interspeech, title = {{The 2020 Personalized Voice Trigger Challenge: Open Datasets, Evaluation Metrics, Baseline System and Results}}, author = {Yan Jia and Xingming Wang and Xiaoyi Qin and Yinping Zhang and Xuyang Wang and Junjie Wang and Dong Zhang and Ming Li}, year = {2021},...
The 2020 Personalized Voice Trigger Challenge (PVTC2020) addresses two different research problems in a unified setup: joint wake-up word detection with speaker verification on close-talking single microphone data and far-field multi-channel microphone array data. Specially, the second task poses an additional cross-ch...
2101.01935
title_judge
wang21ea_interspeech
Auto-KWS 2021 Challenge: Task, Datasets, and Baselines
[ "Jingsong Wang", "Yuxuan He", "Chunyu Zhao", "Qijie Shao", "Wei-Wei Tu", "Tom Ko", "Hung-yi Lee", "Lei Xie" ]
https://www.isca-archive.org/interspeech_2021/wang21ea_interspeech.html
https://www.isca-archive.org/interspeech_2021/wang21ea_interspeech.pdf
10.21437/Interspeech.2021-817
4244-4248
@inproceedings{wang21ea_interspeech, title = {{Auto-KWS 2021 Challenge: Task, Datasets, and Baselines}}, author = {Jingsong Wang and Yuxuan He and Chunyu Zhao and Qijie Shao and Wei-Wei Tu and Tom Ko and Hung-yi Lee and Lei Xie}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4244--...
Auto-KWS 2021 challenge calls for automated machine learning (AutoML) solutions to automate the process of applying machine learning to a customized keyword spotting task. Compared with other keyword spotting tasks, Auto-KWS challenge has the following three characteristics: 1) The challenge focuses on the problem of c...
2104.00513
title_snapshot
berg21_interspeech
Keyword Transformer: A Self-Attention Model for Keyword Spotting
[ "Axel Berg", "Mark O’Connor", "Miguel Tairum Cruz" ]
https://www.isca-archive.org/interspeech_2021/berg21_interspeech.html
https://www.isca-archive.org/interspeech_2021/berg21_interspeech.pdf
10.21437/Interspeech.2021-1286
4249-4253
@inproceedings{berg21_interspeech, title = {{Keyword Transformer: A Self-Attention Model for Keyword Spotting}}, author = {Axel Berg and Mark O’Connor and Miguel Tairum Cruz}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4249--4253}, doi = {10.21437/Interspeech.2021-1286},...
The Transformer architecture has been successful across many domains, including natural language processing, computer vision and speech recognition. In keyword spotting, self-attention has primarily been used on top of convolutional or recurrent encoders. We investigate a range of ways to adapt the Transformer architec...
2104.00769
title_snapshot
awasthi21_interspeech
Teaching Keyword Spotters to Spot New Keywords with Limited Examples
[ "Abhijeet Awasthi", "Kevin Kilgour", "Hassan Rom" ]
https://www.isca-archive.org/interspeech_2021/awasthi21_interspeech.html
https://www.isca-archive.org/interspeech_2021/awasthi21_interspeech.pdf
10.21437/Interspeech.2021-1395
4254-4258
@inproceedings{awasthi21_interspeech, title = {{Teaching Keyword Spotters to Spot New Keywords with Limited Examples}}, author = {Abhijeet Awasthi and Kevin Kilgour and Hassan Rom}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4254--4258}, doi = {10.21437/Interspeech.2021-...
Learning to recognize new keywords with just a few examples is essential for personalizing keyword spotting (KWS) models to a user’s choice of keywords. However, modern KWS models are typically trained on large datasets and restricted to a small vocabulary of keywords, limiting their transferability to a broad range of...
2106.02443
title_snapshot
wang21fa_interspeech
A Comparative Study on Recent Neural Spoofing Countermeasures for Synthetic Speech Detection
[ "Xin Wang", "Junichi Yamagishi" ]
https://www.isca-archive.org/interspeech_2021/wang21fa_interspeech.html
https://www.isca-archive.org/interspeech_2021/wang21fa_interspeech.pdf
10.21437/Interspeech.2021-702
4259-4263
@inproceedings{wang21fa_interspeech, title = {{A Comparative Study on Recent Neural Spoofing Countermeasures for Synthetic Speech Detection}}, author = {Xin Wang and Junichi Yamagishi}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4259--4263}, doi = {10.21437/Interspeech.2...
A great deal of recent research effort on speech spoofing countermeasures has been invested into back-end neural networks and training criteria. We contribute to this effort with a comparative perspective in this study. Our comparison of countermeasure models on the ASVspoof 2019 logical access scenario takes into acco...
2103.11326
title_snapshot
zhang21ca_interspeech
An Initial Investigation for Detecting Partially Spoofed Audio
[ "Lin Zhang", "Xin Wang", "Erica Cooper", "Junichi Yamagishi", "Jose Patino", "Nicholas Evans" ]
https://www.isca-archive.org/interspeech_2021/zhang21ca_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhang21ca_interspeech.pdf
10.21437/Interspeech.2021-738
4264-4268
@inproceedings{zhang21ca_interspeech, title = {{An Initial Investigation for Detecting Partially Spoofed Audio}}, author = {Lin Zhang and Xin Wang and Erica Cooper and Junichi Yamagishi and Jose Patino and Nicholas Evans}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4264--4268}, ...
All existing databases of spoofed speech contain attack data that is spoofed in its entirety. In practice, it is entirely plausible that successful attacks can be mounted with utterances that are only partially spoofed. By definition, partially-spoofed utterances contain a mix of both spoofed and bona fide segments, wh...
2104.02518
title_snapshot
xie21_interspeech
Siamese Network with wav2vec Feature for Spoofing Speech Detection
[ "Yang Xie", "Zhenchuan Zhang", "Yingchun Yang" ]
https://www.isca-archive.org/interspeech_2021/xie21_interspeech.html
https://www.isca-archive.org/interspeech_2021/xie21_interspeech.pdf
10.21437/Interspeech.2021-847
4269-4273
@inproceedings{xie21_interspeech, title = {{Siamese Network with wav2vec Feature for Spoofing Speech Detection}}, author = {Yang Xie and Zhenchuan Zhang and Yingchun Yang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4269--4273}, doi = {10.21437/Interspeech.2021-847}, i...
Automatic speaker verification is vulnerable to spoofing attacks with synthesized or converted speech. Although high-performance anti-spoofing countermeasures can achieve high accuracy when the training and testing spoofing attack examples are similarly distributed, their performance degrades significantly when confron...
null
null
cheng21b_interspeech
Cross-Database Replay Detection in Terminal-Dependent Speaker Verification
[ "Xingliang Cheng", "Mingxing Xu", "Thomas Fang Zheng" ]
https://www.isca-archive.org/interspeech_2021/cheng21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/cheng21b_interspeech.pdf
10.21437/Interspeech.2021-960
4274-4278
@inproceedings{cheng21b_interspeech, title = {{Cross-Database Replay Detection in Terminal-Dependent Speaker Verification}}, author = {Xingliang Cheng and Mingxing Xu and Thomas Fang Zheng}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4274--4278}, doi = {10.21437/Interspe...
The vulnerability of automatic speaker verification (ASV) systems against replay attacks becomes a severe problem. Although various methods have been proposed for replay detection, the generalization capability is still limited. For instance, a detection model trained on one database may fully fail when tested on anoth...
null
null
zhang21da_interspeech
The Effect of Silence and Dual-Band Fusion in Anti-Spoofing System
[ "Yuxiang Zhang", "Wenchao Wang", "Pengyuan Zhang" ]
https://www.isca-archive.org/interspeech_2021/zhang21da_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhang21da_interspeech.pdf
10.21437/Interspeech.2021-1281
4279-4283
@inproceedings{zhang21da_interspeech, title = {{The Effect of Silence and Dual-Band Fusion in Anti-Spoofing System}}, author = {Yuxiang Zhang and Wenchao Wang and Pengyuan Zhang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4279--4283}, doi = {10.21437/Interspeech.2021-12...
The current neural network based anti-spoofing systems have poor robustness. Their performance degrades further after voice activity detection (VAD) performed, making it difficult to be applied in practice. This work investigated the effect of silence at the beginning and end of speech, finding that silent differences ...
null
null
peng21d_interspeech
Pairing Weak with Strong: Twin Models for Defending Against Adversarial Attack on Speaker Verification
[ "Zhiyuan Peng", "Xu Li", "Tan Lee" ]
https://www.isca-archive.org/interspeech_2021/peng21d_interspeech.html
https://www.isca-archive.org/interspeech_2021/peng21d_interspeech.pdf
10.21437/Interspeech.2021-1343
4284-4288
@inproceedings{peng21d_interspeech, title = {{Pairing Weak with Strong: Twin Models for Defending Against Adversarial Attack on Speaker Verification}}, author = {Zhiyuan Peng and Xu Li and Tan Lee}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4284--4288}, doi = {10.21437/...
Vulnerability of speaker verification (SV) systems under adversarial attack receives wide attention recently. Simple and effective countermeasures against such attack are yet to be developed. This paper formulates the task of adversarial defense as a problem of attack detection. The detection is made possible with the ...
null
null
ling21_interspeech
Attention-Based Convolutional Neural Network for ASV Spoofing Detection
[ "Hefei Ling", "Leichao Huang", "Junrui Huang", "Baiyan Zhang", "Ping Li" ]
https://www.isca-archive.org/interspeech_2021/ling21_interspeech.html
https://www.isca-archive.org/interspeech_2021/ling21_interspeech.pdf
10.21437/Interspeech.2021-1404
4289-4293
@inproceedings{ling21_interspeech, title = {{Attention-Based Convolutional Neural Network for ASV Spoofing Detection}}, author = {Hefei Ling and Leichao Huang and Junrui Huang and Baiyan Zhang and Ping Li}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4289--4293}, doi = {1...
In recent years, automatic speaker verification (ASV) algorithms have undergone significant progress. They have been widely deployed in different applications, but the ASV systems are vulnerable to spoofing attacks, such as impersonation, replay, text-to-speech, voice conversion and the recently emerged adversarial att...
null
null
wu21i_interspeech
Voting for the Right Answer: Adversarial Defense for Speaker Verification
[ "Haibin Wu", "Yang Zhang", "Zhiyong Wu", "Dong Wang", "Hung-yi Lee" ]
https://www.isca-archive.org/interspeech_2021/wu21i_interspeech.html
https://www.isca-archive.org/interspeech_2021/wu21i_interspeech.pdf
10.21437/Interspeech.2021-1452
4294-4298
@inproceedings{wu21i_interspeech, title = {{Voting for the Right Answer: Adversarial Defense for Speaker Verification}}, author = {Haibin Wu and Yang Zhang and Zhiyong Wu and Dong Wang and Hung-yi Lee}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4294--4298}, doi = {10.21...
Automatic speaker verification (ASV) is a well developed technology for biometric identification, and has been ubiquitous implemented in security-critic applications, such as banking and access control. However, previous works have shown that ASV is under the radar of adversarial attacks, which are very similar to thei...
2106.07868
title_snapshot
kinnunen21_interspeech
Visualizing Classifier Adjacency Relations: A Case Study in Speaker Verification and Voice Anti-Spoofing
[ "Tomi Kinnunen", "Andreas Nautsch", "Md. Sahidullah", "Nicholas Evans", "Xin Wang", "Massimiliano Todisco", "Héctor Delgado", "Junichi Yamagishi", "Kong Aik Lee" ]
https://www.isca-archive.org/interspeech_2021/kinnunen21_interspeech.html
https://www.isca-archive.org/interspeech_2021/kinnunen21_interspeech.pdf
10.21437/Interspeech.2021-1522
4299-4303
@inproceedings{kinnunen21_interspeech, title = {{Visualizing Classifier Adjacency Relations: A Case Study in Speaker Verification and Voice Anti-Spoofing}}, author = {Tomi Kinnunen and Andreas Nautsch and Md. Sahidullah and Nicholas Evans and Xin Wang and Massimiliano Todisco and Héctor Delgado and Junichi Y...
Whether it be for results summarization, or the analysis of classifier fusion, some means to compare different classifiers can often provide illuminating insight into their behaviour, (dis)similarity or complementarity. We propose a simple method to derive 2D representation from detection scores produced by an arbitrar...
2106.06362
title_snapshot
villalba21_interspeech
Representation Learning to Classify and Detect Adversarial Attacks Against Speaker and Speech Recognition Systems
[ "Jesús Villalba", "Sonal Joshi", "Piotr Żelasko", "Najim Dehak" ]
https://www.isca-archive.org/interspeech_2021/villalba21_interspeech.html
https://www.isca-archive.org/interspeech_2021/villalba21_interspeech.pdf
10.21437/Interspeech.2021-1759
4304-4308
@inproceedings{villalba21_interspeech, title = {{Representation Learning to Classify and Detect Adversarial Attacks Against Speaker and Speech Recognition Systems}}, author = {Jesús Villalba and Sonal Joshi and Piotr Żelasko and Najim Dehak}, year = {2021}, booktitle = {{Interspeech 2021}}, pages ...
Adversarial attacks have become a major threat for machine learning applications. There is a growing interest in studying these attacks in the audio domain, e.g, speech and speaker recognition; and find defenses against them. In this work, we focus on using representation learning to classify/detect attacks w.r.t. the ...
2107.04448
title_snapshot
zhang21ea_interspeech
An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems
[ "You Zhang", "Ge Zhu", "Fei Jiang", "Zhiyao Duan" ]
https://www.isca-archive.org/interspeech_2021/zhang21ea_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhang21ea_interspeech.pdf
10.21437/Interspeech.2021-1820
4309-4313
@inproceedings{zhang21ea_interspeech, title = {{An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems}}, author = {You Zhang and Ge Zhu and Fei Jiang and Zhiyao Duan}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4309--4313}, doi = {10.2...
Spoofing countermeasure (CM) systems are critical in speaker verification; they aim to discern spoofing attacks from bona fide speech trials. In practice, however, acoustic condition variability in speech utterances may significantly degrade the performance of CM systems. In this paper, we conduct a cross-dataset study...
2104.01320
title_snapshot
li21o_interspeech
Channel-Wise Gated Res2Net: Towards Robust Detection of Synthetic Speech Attacks
[ "Xu Li", "Xixin Wu", "Hui Lu", "Xunying Liu", "Helen Meng" ]
https://www.isca-archive.org/interspeech_2021/li21o_interspeech.html
https://www.isca-archive.org/interspeech_2021/li21o_interspeech.pdf
10.21437/Interspeech.2021-2125
4314-4318
@inproceedings{li21o_interspeech, title = {{Channel-Wise Gated Res2Net: Towards Robust Detection of Synthetic Speech Attacks}}, author = {Xu Li and Xixin Wu and Hui Lu and Xunying Liu and Helen Meng}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4314--4318}, doi = {10.2143...
Existing approaches for anti-spoofing in automatic speaker verification (ASV) still lack generalizability to unseen attacks. The Res2Net approach designs a residual-like connection between feature groups within one block, which increases the possible receptive fields and improves the system’s detection generalizability...
2107.08803
title_snapshot
ge21c_interspeech
Partially-Connected Differentiable Architecture Search for Deepfake and Spoofing Detection
[ "Wanying Ge", "Michele Panariello", "Jose Patino", "Massimiliano Todisco", "Nicholas Evans" ]
https://www.isca-archive.org/interspeech_2021/ge21c_interspeech.html
https://www.isca-archive.org/interspeech_2021/ge21c_interspeech.pdf
10.21437/Interspeech.2021-1187
4319-4323
@inproceedings{ge21c_interspeech, title = {{Partially-Connected Differentiable Architecture Search for Deepfake and Spoofing Detection}}, author = {Wanying Ge and Michele Panariello and Jose Patino and Massimiliano Todisco and Nicholas Evans}, year = {2021}, booktitle = {{Interspeech 2021}}, pages...
This paper reports the first successful application of a differentiable architecture search (DARTS) approach to the deepfake and spoofing detection problems. An example of neural architecture search, DARTS operates upon a continuous, differentiable search space which enables both the architecture and parameters to be o...
2104.03123
title_snapshot
peterson21_interspeech
OpenASR20: An Open Challenge for Automatic Speech Recognition of Conversational Telephone Speech in Low-Resource Languages
[ "Kay Peterson", "Audrey Tong", "Yan Yu" ]
https://www.isca-archive.org/interspeech_2021/peterson21_interspeech.html
https://www.isca-archive.org/interspeech_2021/peterson21_interspeech.pdf
10.21437/Interspeech.2021-1930
4324-4328
@inproceedings{peterson21_interspeech, title = {{OpenASR20: An Open Challenge for Automatic Speech Recognition of Conversational Telephone Speech in Low-Resource Languages}}, author = {Kay Peterson and Audrey Tong and Yan Yu}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4324--432...
In 2020, the National Institute of Standards and Technology (NIST), in cooperation with the Intelligence Advanced Research Project Activity (IARPA), conducted an open challenge on automatic speech recognition (ASR) technology for low-resource languages on a challenging data type — conversational telephone speech. The O...
null
null
madikeri21_interspeech
Multitask Adaptation with Lattice-Free MMI for Multi-Genre Speech Recognition of Low Resource Languages
[ "Srikanth Madikeri", "Petr Motlicek", "Hervé Bourlard" ]
https://www.isca-archive.org/interspeech_2021/madikeri21_interspeech.html
https://www.isca-archive.org/interspeech_2021/madikeri21_interspeech.pdf
10.21437/Interspeech.2021-1778
4329-4333
@inproceedings{madikeri21_interspeech, title = {{Multitask Adaptation with Lattice-Free MMI for Multi-Genre Speech Recognition of Low Resource Languages}}, author = {Srikanth Madikeri and Petr Motlicek and Hervé Bourlard}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4329--4333}, ...
In this paper, we develop Automatic Speech Recognition (ASR) systems for multi-genre speech recognition of low-resource languages where training data is predominantly conversational speech but test data can be in one of the following genres: news broadcast, topical broadcast and conversational speech. ASR for low-resou...
null
null
zhu21f_interspeech
An Improved Wav2Vec 2.0 Pre-Training Approach Using Enhanced Local Dependency Modeling for Speech Recognition
[ "Qiu-shi Zhu", "Jie Zhang", "Ming-hui Wu", "Xin Fang", "Li-Rong Dai" ]
https://www.isca-archive.org/interspeech_2021/zhu21f_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhu21f_interspeech.pdf
10.21437/Interspeech.2021-67
4334-4338
@inproceedings{zhu21f_interspeech, title = {{An Improved Wav2Vec 2.0 Pre-Training Approach Using Enhanced Local Dependency Modeling for Speech Recognition}}, author = {Qiu-shi Zhu and Jie Zhang and Ming-hui Wu and Xin Fang and Li-Rong Dai}, year = {2021}, booktitle = {{Interspeech 2021}}, pages ...
wav2vec 2.0 is a recently proposed self-supervised pre-training framework for learning speech representation. It utilizes a transformer to learn global contextual representation, which is effective especially in low-resource scenarios. Besides, it was shown that combining convolution neural network and transformer to m...
null
null
lin21i_interspeech
Systems for Low-Resource Speech Recognition Tasks in Open Automatic Speech Recognition and Formosa Speech Recognition Challenges
[ "Hung-Pang Lin", "Yu-Jia Zhang", "Chia-Ping Chen" ]
https://www.isca-archive.org/interspeech_2021/lin21i_interspeech.html
https://www.isca-archive.org/interspeech_2021/lin21i_interspeech.pdf
10.21437/Interspeech.2021-358
4339-4343
@inproceedings{lin21i_interspeech, title = {{Systems for Low-Resource Speech Recognition Tasks in Open Automatic Speech Recognition and Formosa Speech Recognition Challenges}}, author = {Hung-Pang Lin and Yu-Jia Zhang and Chia-Ping Chen}, year = {2021}, booktitle = {{Interspeech 2021}}, pages ...
We, in the team name of NSYSU-MITLab, have participated in low-resource speech recognition of the Open Automatic Speech Recognition Challenge 2020 (OpenASR20) and Formosa Speech Recognition Challenge 2020 (FSR-2020). For the tasks in the challenges, we build and compare end-to-end (E2E) systems and Deep Neural Network ...
null
null
zhao21c_interspeech
The TNT Team System Descriptions of Cantonese and Mongolian for IARPA OpenASR20
[ "Jing Zhao", "Zhiqiang Lv", "Ambyera Han", "Guan-Bo Wang", "Guixin Shi", "Jian Kang", "Jinghao Yan", "Pengfei Hu", "Shen Huang", "Wei-Qiang Zhang" ]
https://www.isca-archive.org/interspeech_2021/zhao21c_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhao21c_interspeech.pdf
10.21437/Interspeech.2021-1063
4344-4348
@inproceedings{zhao21c_interspeech, title = {{The TNT Team System Descriptions of Cantonese and Mongolian for IARPA OpenASR20}}, author = {Jing Zhao and Zhiqiang Lv and Ambyera Han and Guan-Bo Wang and Guixin Shi and Jian Kang and Jinghao Yan and Pengfei Hu and Shen Huang and Wei-Qiang Zhang}, year = ...
This paper presents our work for OpenASR20 Challenge. We describe our Automatic Speech Recognition (ASR) systems for Cantonese and Mongolian under both constrained and unconstrained conditions. For constrained condition, a hybrid NN-HMM ASR system play the main role, while for unconstrained condition, an end-to-end ASR...
null
null
alumae21_interspeech
Combining Hybrid and End-to-End Approaches for the OpenASR20 Challenge
[ "Tanel Alumäe", "Jiaming Kong" ]
https://www.isca-archive.org/interspeech_2021/alumae21_interspeech.html
https://www.isca-archive.org/interspeech_2021/alumae21_interspeech.pdf
10.21437/Interspeech.2021-1086
4349-4353
@inproceedings{alumae21_interspeech, title = {{Combining Hybrid and End-to-End Approaches for the OpenASR20 Challenge}}, author = {Tanel Alumäe and Jiaming Kong}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4349--4353}, doi = {10.21437/Interspeech.2021-1086}, issn ...
This paper describes the TalTech team submission to the OpenASR20 Challenge. OpenASR20 evaluated low-resource speech recognition technologies across 10 languages, using only 10 hours of training data in the constrained condition. Our ASR systems used hybrid CNN-TDNNF-based acoustic models, trained with different data a...
null
null
morris21_interspeech
One Size Does Not Fit All in Resource-Constrained ASR
[ "Ethan Morris", "Robbie Jimerson", "Emily Prud’hommeaux" ]
https://www.isca-archive.org/interspeech_2021/morris21_interspeech.html
https://www.isca-archive.org/interspeech_2021/morris21_interspeech.pdf
10.21437/Interspeech.2021-1970
4354-4358
@inproceedings{morris21_interspeech, title = {{One Size Does Not Fit All in Resource-Constrained ASR}}, author = {Ethan Morris and Robbie Jimerson and Emily Prud’hommeaux}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4354--4358}, doi = {10.21437/Interspeech.2021-1970}, ...
The application of deep neural networks to the task of acoustic modeling for automatic speech recognition has resulted in dramatic decreases in ASR word error rates, enabling the use of this technology for interacting with smart phones and personal home assistants in high-resource languages. Developing ASR models of th...
null
null
gimeno21_interspeech
Unsupervised Representation Learning for Speech Activity Detection in the Fearless Steps Challenge 2021
[ "Pablo Gimeno", "Alfonso Ortega", "Antonio Miguel", "Eduardo Lleida" ]
https://www.isca-archive.org/interspeech_2021/gimeno21_interspeech.html
https://www.isca-archive.org/interspeech_2021/gimeno21_interspeech.pdf
10.21437/Interspeech.2021-309
4359-4363
@inproceedings{gimeno21_interspeech, title = {{Unsupervised Representation Learning for Speech Activity Detection in the Fearless Steps Challenge 2021}}, author = {Pablo Gimeno and Alfonso Ortega and Antonio Miguel and Eduardo Lleida}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {...
In this paper, we describe the ViVoLab speech activity detection (SAD) system submitted to the Fearless Steps Challenge Phase III. This series of challenges have proposed a number of speech processing task dealing with audio from Apollo space missions over the last few years. The focus in this edition is set on the gen...
null
null
vuong21_interspeech
The Application of Learnable STRF Kernels to the 2021 Fearless Steps Phase-03 SAD Challenge
[ "Tyler Vuong", "Yangyang Xia", "Richard M. Stern" ]
https://www.isca-archive.org/interspeech_2021/vuong21_interspeech.html
https://www.isca-archive.org/interspeech_2021/vuong21_interspeech.pdf
10.21437/Interspeech.2021-651
4364-4368
@inproceedings{vuong21_interspeech, title = {{The Application of Learnable STRF Kernels to the 2021 Fearless Steps Phase-03 SAD Challenge}}, author = {Tyler Vuong and Yangyang Xia and Richard M. Stern}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4364--4368}, doi = {10.21...
We describe a deep-learning-based system developed for the Fearless Steps Phase-03 Speech Activity Detection (SAD) challenge. The system includes both learnable spectro-temporal receptive fields (STRFs) and unconstrained 2-dimensional convolutional kernels in the first layer. Experiments show that the inclusion of lear...
null
null
sarfjoo21_interspeech
Speech Activity Detection Based on Multilingual Speech Recognition System
[ "Seyyed Saeed Sarfjoo", "Srikanth Madikeri", "Petr Motlicek" ]
https://www.isca-archive.org/interspeech_2021/sarfjoo21_interspeech.html
https://www.isca-archive.org/interspeech_2021/sarfjoo21_interspeech.pdf
10.21437/Interspeech.2021-1058
4369-4373
@inproceedings{sarfjoo21_interspeech, title = {{Speech Activity Detection Based on Multilingual Speech Recognition System}}, author = {Seyyed Saeed Sarfjoo and Srikanth Madikeri and Petr Motlicek}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4369--4373}, doi = {10.21437/I...
To better model the contextual information and increase the generalization ability of the Speech Activity Detection (SAD) system, this paper leverages a multilingual Automatic Speech Recognition (ASR) system to perform SAD. Sequence-discriminative training of Acoustic Model (AM) using Lattice-Free Maximum Mutual Inform...
2010.12277
title_snapshot
luckenbaugh21_interspeech
Voice Activity Detection with Teacher-Student Domain Emulation
[ "Jarrod Luckenbaugh", "Samuel Abplanalp", "Rachel Gonzalez", "Daniel Fulford", "David Gard", "Carlos Busso" ]
https://www.isca-archive.org/interspeech_2021/luckenbaugh21_interspeech.html
https://www.isca-archive.org/interspeech_2021/luckenbaugh21_interspeech.pdf
10.21437/Interspeech.2021-1234
4374-4378
@inproceedings{luckenbaugh21_interspeech, title = {{Voice Activity Detection with Teacher-Student Domain Emulation}}, author = {Jarrod Luckenbaugh and Samuel Abplanalp and Rachel Gonzalez and Daniel Fulford and David Gard and Carlos Busso}, year = {2021}, booktitle = {{Interspeech 2021}}, pages ...
Transfer learning is a promising approach to increase performance for many speech-based systems, including voice activity detection (VAD). Domain adaptation, a subfield of transfer learning, often improves model conditioning in the presence of a mismatch between train-test conditions. This study proposes a formulation ...
null
null
ghahabi21_interspeech
EML Online Speech Activity Detection for the Fearless Steps Challenge Phase-III
[ "Omid Ghahabi", "Volker Fischer" ]
https://www.isca-archive.org/interspeech_2021/ghahabi21_interspeech.html
https://www.isca-archive.org/interspeech_2021/ghahabi21_interspeech.pdf
10.21437/Interspeech.2021-1456
4379-4382
@inproceedings{ghahabi21_interspeech, title = {{EML Online Speech Activity Detection for the Fearless Steps Challenge Phase-III}}, author = {Omid Ghahabi and Volker Fischer}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4379--4382}, doi = {10.21437/Interspeech.2021-1456}, ...
Speech Activity Detection (SAD), locating speech segments within an audio recording, is a main part of most speech technology applications. Robust SAD is usually more difficult in noisy conditions with varying signal-to-noise ratios (SNR). The Fearless Steps challenge has recently provided such data from the NASA Apoll...
2106.11075
title_snapshot
opatka21_interspeech
Device Playback Augmentation with Echo Cancellation for Keyword Spotting
[ "Kuba Łopatka", "Katarzyna Kaszuba-Miotke", "Piotr Klinke", "Paweł Trella" ]
https://www.isca-archive.org/interspeech_2021/opatka21_interspeech.html
https://www.isca-archive.org/interspeech_2021/opatka21_interspeech.pdf
10.21437/Interspeech.2021-1316
4383-4387
@inproceedings{opatka21_interspeech, title = {{Device Playback Augmentation with Echo Cancellation for Keyword Spotting}}, author = {Kuba Łopatka and Katarzyna Kaszuba-Miotke and Piotr Klinke and Paweł Trella}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4383--4387}, doi ...
Keyword spotting (KWS) is required to operate in device playback conditions in which the device itself plays interfering signals. We propose a new method to augment the training set and adapt the acoustic model to the playback environment. It is based on acoustic simulation which models the coupling between the device’...
null
null
yusuf21_interspeech
End-to-End Open Vocabulary Keyword Search
[ "Bolaji Yusuf", "Alican Gok", "Batuhan Gundogdu", "Murat Saraclar" ]
https://www.isca-archive.org/interspeech_2021/yusuf21_interspeech.html
https://www.isca-archive.org/interspeech_2021/yusuf21_interspeech.pdf
10.21437/Interspeech.2021-1399
4388-4392
@inproceedings{yusuf21_interspeech, title = {{End-to-End Open Vocabulary Keyword Search}}, author = {Bolaji Yusuf and Alican Gok and Batuhan Gundogdu and Murat Saraclar}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4388--4392}, doi = {10.21437/Interspeech.2021-1399}, is...
Recently, neural approaches to spoken content retrieval have become popular. However, they tend to be restricted in their vocabulary or in their ability to deal with imbalanced test settings. These restrictions limit their applicability in keyword search, where the set of queries is not known beforehand, and where the ...
2108.10357
title_snapshot
merkx21_interspeech
Semantic Sentence Similarity: Size does not Always Matter
[ "Danny Merkx", "Stefan L. Frank", "Mirjam Ernestus" ]
https://www.isca-archive.org/interspeech_2021/merkx21_interspeech.html
https://www.isca-archive.org/interspeech_2021/merkx21_interspeech.pdf
10.21437/Interspeech.2021-1464
4393-4397
@inproceedings{merkx21_interspeech, title = {{Semantic Sentence Similarity: Size does not Always Matter}}, author = {Danny Merkx and Stefan L. Frank and Mirjam Ernestus}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4393--4397}, doi = {10.21437/Interspeech.2021-1464}, is...
This study addresses the question whether visually grounded speech recognition (VGS) models learn to capture sentence semantics without access to any prior linguistic knowledge. We produce synthetic and natural spoken versions of a well known semantic textual similarity database and show that our VGS model produces emb...
2106.08648
title_snapshot
svec21_interspeech
Spoken Term Detection and Relevance Score Estimation Using Dot-Product of Pronunciation Embeddings
[ "Jan Švec", "Luboš Šmídl", "Josef V. Psutka", "Aleš Pražák" ]
https://www.isca-archive.org/interspeech_2021/svec21_interspeech.html
https://www.isca-archive.org/interspeech_2021/svec21_interspeech.pdf
10.21437/Interspeech.2021-1704
4398-4402
@inproceedings{svec21_interspeech, title = {{Spoken Term Detection and Relevance Score Estimation Using Dot-Product of Pronunciation Embeddings}}, author = {Jan Švec and Luboš Šmídl and Josef V. Psutka and Aleš Pražák}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4398--4402}, d...
The paper describes a novel approach to Spoken Term Detection (STD) in large spoken archives using deep LSTM networks. The work is based on the previous approach of using Siamese neural networks for STD and naturally extends it to directly localize a spoken term and estimate its relevance score. The phoneme confusion n...
2210.11895
title_snapshot
buet21_interspeech
Toward Genre Adapted Closed Captioning
[ "François Buet", "François Yvon" ]
https://www.isca-archive.org/interspeech_2021/buet21_interspeech.html
https://www.isca-archive.org/interspeech_2021/buet21_interspeech.pdf
10.21437/Interspeech.2021-1762
4403-4407
@inproceedings{buet21_interspeech, title = {{Toward Genre Adapted Closed Captioning}}, author = {François Buet and François Yvon}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4403--4407}, doi = {10.21437/Interspeech.2021-1762}, issn = {2958-1796}, }
This paper studies the generation of intralingual closed captions from automatic speech transcripts, with the aim to assess techniques for multi-genre captioning. Captions and subtitles greatly vary in form and content depending on the programs genres and subtitling styles, resulting for instance in significantly diffe...
null
null
korzekwa21b_interspeech
Weakly-Supervised Word-Level Pronunciation Error Detection in Non-Native English Speech
[ "Daniel Korzekwa", "Jaime Lorenzo-Trueba", "Thomas Drugman", "Shira Calamaro", "Bozena Kostek" ]
https://www.isca-archive.org/interspeech_2021/korzekwa21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/korzekwa21b_interspeech.pdf
10.21437/Interspeech.2021-38
4408-4412
@inproceedings{korzekwa21b_interspeech, title = {{Weakly-Supervised Word-Level Pronunciation Error Detection in Non-Native English Speech}}, author = {Daniel Korzekwa and Jaime Lorenzo-Trueba and Thomas Drugman and Shira Calamaro and Bozena Kostek}, year = {2021}, booktitle = {{Interspeech 2021}}, ...
We propose a weakly-supervised model for word-level mispronunciation detection in non-native (L2) English speech. To train this model, phonetically transcribed L2 speech is not required and we only need to mark mispronounced words. The lack of phonetic transcriptions for L2 speech means that the model has to learn only...
2106.03494
title_snapshot
kanda21b_interspeech
End-to-End Speaker-Attributed ASR with Transformer
[ "Naoyuki Kanda", "Guoli Ye", "Yashesh Gaur", "Xiaofei Wang", "Zhong Meng", "Zhuo Chen", "Takuya Yoshioka" ]
https://www.isca-archive.org/interspeech_2021/kanda21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/kanda21b_interspeech.pdf
10.21437/Interspeech.2021-101
4413-4417
@inproceedings{kanda21b_interspeech, title = {{End-to-End Speaker-Attributed ASR with Transformer}}, author = {Naoyuki Kanda and Guoli Ye and Yashesh Gaur and Xiaofei Wang and Zhong Meng and Zhuo Chen and Takuya Yoshioka}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4413--4417}, ...
This paper presents our recent effort on end-to-end speaker-attributed automatic speech recognition, which jointly performs speaker counting, speech recognition and speaker identification for monaural multi-talker audio. Firstly, we thoroughly update the model architecture that was previously designed based on a long s...
2104.02128
title_snapshot
soltau21_interspeech
Understanding Medical Conversations: Rich Transcription, Confidence Scores & Information Extraction
[ "Hagen Soltau", "Mingqiu Wang", "Izhak Shafran", "Laurent El Shafey" ]
https://www.isca-archive.org/interspeech_2021/soltau21_interspeech.html
https://www.isca-archive.org/interspeech_2021/soltau21_interspeech.pdf
10.21437/Interspeech.2021-691
4418-4422
@inproceedings{soltau21_interspeech, title = {{Understanding Medical Conversations: Rich Transcription, Confidence Scores & Information Extraction}}, author = {Hagen Soltau and Mingqiu Wang and Izhak Shafran and Laurent El Shafey}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4418...
In this paper, we describe novel components for extracting clinically relevant information from medical conversations which will be available as Google APIs. We describe a transformer-based Recurrent Neural Network Transducer (RNN-T) model tailored for long-form audio, which can produce rich transcriptions including sp...
2104.02219
title_snapshot
vidal21_interspeech
Phone-Level Pronunciation Scoring for Spanish Speakers Learning English Using a GOP-DNN System
[ "Jazmín Vidal", "Cyntia Bonomi", "Marcelo Sancinetti", "Luciana Ferrer" ]
https://www.isca-archive.org/interspeech_2021/vidal21_interspeech.html
https://www.isca-archive.org/interspeech_2021/vidal21_interspeech.pdf
10.21437/Interspeech.2021-745
4423-4427
@inproceedings{vidal21_interspeech, title = {{Phone-Level Pronunciation Scoring for Spanish Speakers Learning English Using a GOP-DNN System}}, author = {Jazmín Vidal and Cyntia Bonomi and Marcelo Sancinetti and Luciana Ferrer}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4423--4...
In today’s globalized world being able to communicate in English is crucial to many people. Computer assisted pronunciation training (CAPT) systems can help students achieve English proficiency by providing an accessible way to practice, offering personalized feedback. However, phone-level pronunciation scoring is stil...
null
null
xu21k_interspeech
Explore wav2vec 2.0 for Mispronunciation Detection
[ "Xiaoshuo Xu", "Yueteng Kang", "Songjun Cao", "Binghuai Lin", "Long Ma" ]
https://www.isca-archive.org/interspeech_2021/xu21k_interspeech.html
https://www.isca-archive.org/interspeech_2021/xu21k_interspeech.pdf
10.21437/Interspeech.2021-777
4428-4432
@inproceedings{xu21k_interspeech, title = {{Explore wav2vec 2.0 for Mispronunciation Detection}}, author = {Xiaoshuo Xu and Yueteng Kang and Songjun Cao and Binghuai Lin and Long Ma}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4428--4432}, doi = {10.21437/Interspeech.202...
This paper presents an initial attempt to use self-supervised learning for Mispronunciation Detection. Unlike existing methods that use speech recognition corpus to train models, we exploit unlabeled data and utilize a self-supervised learning technique, Wav2vec 2.0, for pretraining. After the pretraining process, the ...
null
null
ando21_interspeech
Lexical Density Analysis of Word Productions in Japanese English Using Acoustic Word Embeddings
[ "Shintaro Ando", "Nobuaki Minematsu", "Daisuke Saito" ]
https://www.isca-archive.org/interspeech_2021/ando21_interspeech.html
https://www.isca-archive.org/interspeech_2021/ando21_interspeech.pdf
10.21437/Interspeech.2021-853
4433-4437
@inproceedings{ando21_interspeech, title = {{Lexical Density Analysis of Word Productions in Japanese English Using Acoustic Word Embeddings}}, author = {Shintaro Ando and Nobuaki Minematsu and Daisuke Saito}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4433--4437}, doi =...
In L2 pronunciation, what kind of phonetic errors are more influential to intelligibility reduction? Teachers say that learners’ utterances become unintelligible when words are pronounced with such errors that make the words misidentified as others. In this paper, we focus on Japanese English (JE), where the number of ...
null
null
lin21j_interspeech
Deep Feature Transfer Learning for Automatic Pronunciation Assessment
[ "Binghuai Lin", "Liyuan Wang" ]
https://www.isca-archive.org/interspeech_2021/lin21j_interspeech.html
https://www.isca-archive.org/interspeech_2021/lin21j_interspeech.pdf
10.21437/Interspeech.2021-931
4438-4442
@inproceedings{lin21j_interspeech, title = {{Deep Feature Transfer Learning for Automatic Pronunciation Assessment}}, author = {Binghuai Lin and Liyuan Wang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4438--4442}, doi = {10.21437/Interspeech.2021-931}, issn = {29...
Automatic pronunciation assessment is commonly developed to evaluate pronunciation quality of second language (L2) learners. Traditional methods for automatic pronunciation assessment normally utilize speech features such as Goodness of pronunciation (GOP), which may not provide sufficient information for the pronuncia...
null
null
zhang21fa_interspeech
Multilingual Speech Evaluation: Case Studies on English, Malay and Tamil
[ "Huayun Zhang", "Ke Shi", "Nancy F. Chen" ]
https://www.isca-archive.org/interspeech_2021/zhang21fa_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhang21fa_interspeech.pdf
10.21437/Interspeech.2021-1258
4443-4447
@inproceedings{zhang21fa_interspeech, title = {{Multilingual Speech Evaluation: Case Studies on English, Malay and Tamil}}, author = {Huayun Zhang and Ke Shi and Nancy F. Chen}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4443--4447}, doi = {10.21437/Interspeech.2021-1258...
Speech evaluation is an essential component in computer-assisted language learning (CALL). While speech evaluation on English has been popular, automatic speech scoring on low resource languages remains challenging. Work in this area has focused on monolingual specific designs and handcrafted features stemming from res...
2107.03675
title_snapshot
peng21e_interspeech
A Study on Fine-Tuning wav2vec2.0 Model for the Task of Mispronunciation Detection and Diagnosis
[ "Linkai Peng", "Kaiqi Fu", "Binghuai Lin", "Dengfeng Ke", "Jinsong Zhan" ]
https://www.isca-archive.org/interspeech_2021/peng21e_interspeech.html
https://www.isca-archive.org/interspeech_2021/peng21e_interspeech.pdf
10.21437/Interspeech.2021-1344
4448-4452
@inproceedings{peng21e_interspeech, title = {{A Study on Fine-Tuning wav2vec2.0 Model for the Task of Mispronunciation Detection and Diagnosis}}, author = {Linkai Peng and Kaiqi Fu and Binghuai Lin and Dengfeng Ke and Jinsong Zhan}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {444...
Mispronunciation detection and diagnosis (MDD) technology is a key component of computer-assisted pronunciation training system (CAPT). The mainstream method is based on deep neural network automatic speech recognition. Unfortunately, the technique requires massive human-annotated speech recordings for training. Due to...
null
null
qiao21b_interspeech
The Impact of ASR on the Automatic Analysis of Linguistic Complexity and Sophistication in Spontaneous L2 Speech
[ "Yu Qiao", "Wei Zhou", "Elma Kerz", "Ralf Schlüter" ]
https://www.isca-archive.org/interspeech_2021/qiao21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/qiao21b_interspeech.pdf
10.21437/Interspeech.2021-1402
4453-4457
@inproceedings{qiao21b_interspeech, title = {{The Impact of ASR on the Automatic Analysis of Linguistic Complexity and Sophistication in Spontaneous L2 Speech}}, author = {Yu Qiao and Wei Zhou and Elma Kerz and Ralf Schlüter}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4453--445...
In recent years, automated approaches to assessing linguistic complexity in second language (L2) writing have made significant progress in gauging learner performance, predicting human ratings of the quality of learner productions, and benchmarking L2 development. In contrast, there is comparatively little work in the ...
2104.08529
title_snapshot
tanaka21c_interspeech
End-to-End Rich Transcription-Style Automatic Speech Recognition with Semi-Supervised Learning
[ "Tomohiro Tanaka", "Ryo Masumura", "Mana Ihori", "Akihiko Takashima", "Shota Orihashi", "Naoki Makishima" ]
https://www.isca-archive.org/interspeech_2021/tanaka21c_interspeech.html
https://www.isca-archive.org/interspeech_2021/tanaka21c_interspeech.pdf
10.21437/Interspeech.2021-1981
4458-4462
@inproceedings{tanaka21c_interspeech, title = {{End-to-End Rich Transcription-Style Automatic Speech Recognition with Semi-Supervised Learning}}, author = {Tomohiro Tanaka and Ryo Masumura and Mana Ihori and Akihiko Takashima and Shota Orihashi and Naoki Makishima}, year = {2021}, booktitle = {{Inte...
We propose a semi-supervised learning method for building end-to-end rich transcription-style automatic speech recognition (RT-ASR) systems from small-scale rich transcription-style and large-scale common transcription-style datasets. In spontaneous speech tasks, various speech phenomena such as fillers, word fragments...
2107.05382
title_snapshot
cumbal21_interspeech
“You don’t understand me!”: Comparing ASR Results for L1 and L2 Speakers of Swedish
[ "Ronald Cumbal", "Birger Moell", "José Lopes", "Olov Engwall" ]
https://www.isca-archive.org/interspeech_2021/cumbal21_interspeech.html
https://www.isca-archive.org/interspeech_2021/cumbal21_interspeech.pdf
10.21437/Interspeech.2021-2140
4463-4467
@inproceedings{cumbal21_interspeech, title = {{“You don’t understand me!”: Comparing ASR Results for L1 and L2 Speakers of Swedish}}, author = {Ronald Cumbal and Birger Moell and José Lopes and Olov Engwall}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4463--4467}, doi = ...
The performance of Automatic Speech Recognition (ASR) systems has constantly increased in state-of-the-art development. However, performance tends to decrease considerably in more challenging conditions (e.g., background noise, multiple speaker social conversations) and with more atypical speakers (e.g., children, non-...
2405.13379
title_snapshot
zhang21ga_interspeech
NeMo Inverse Text Normalization: From Development to Production
[ "Yang Zhang", "Evelina Bakhturina", "Kyle Gorman", "Boris Ginsburg" ]
https://www.isca-archive.org/interspeech_2021/zhang21ga_interspeech.html
https://www.isca-archive.org/interspeech_2021/zhang21ga_interspeech.pdf
10.21437/Interspeech.2021-1571
4468-4472
@inproceedings{zhang21ga_interspeech, title = {{NeMo Inverse Text Normalization: From Development to Production}}, author = {Yang Zhang and Evelina Bakhturina and Kyle Gorman and Boris Ginsburg}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4468--4472}, doi = {10.21437/Int...
Inverse text normalization (ITN) converts spoken-domain automatic speech recognition (ASR) output into written-domain text to improve the readability of the ASR output. Many state-of-the-art ITN systems use hand-written weighted finite-state transducer (WFST) grammars since this task has extremely low tolerance to unre...
2104.05055
title_snapshot
naijo21_interspeech
Improvement of Automatic English Pronunciation Assessment with Small Number of Utterances Using Sentence Speakability
[ "Satsuki Naijo", "Akinori Ito", "Takashi Nose" ]
https://www.isca-archive.org/interspeech_2021/naijo21_interspeech.html
https://www.isca-archive.org/interspeech_2021/naijo21_interspeech.pdf
10.21437/Interspeech.2021-1132
4473-4477
@inproceedings{naijo21_interspeech, title = {{Improvement of Automatic English Pronunciation Assessment with Small Number of Utterances Using Sentence Speakability}}, author = {Satsuki Naijo and Akinori Ito and Takashi Nose}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4473--4477...
The current Computer-Assisted Pronunciation Training (CAPT) system uses DNN-based speech recognition results to evaluate learner’s pronunciation with high accuracy when using many utterances for the evaluation. However, when we use only a few utterances, the accuracy of the CAPT system deteriorates. One reason for the ...
null
null
haider21_interspeech
Affect Recognition Through Scalogram and Multi-Resolution Cochleagram Features
[ "Fasih Haider", "Saturnino Luz" ]
https://www.isca-archive.org/interspeech_2021/haider21_interspeech.html
https://www.isca-archive.org/interspeech_2021/haider21_interspeech.pdf
10.21437/Interspeech.2021-1761
4478-4482
@inproceedings{haider21_interspeech, title = {{Affect Recognition Through Scalogram and Multi-Resolution Cochleagram Features}}, author = {Fasih Haider and Saturnino Luz}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4478--4482}, doi = {10.21437/Interspeech.2021-1761}, i...
An approach to the categorization of voice samples according to emotions expressed by the speaker is proposed which uses Multi-Resolution Cochleagram (MRCG) and scalogram features in a novel way. Audio recordings from the EmoDB, EMOVO and Savee Data-sets are employed in training and testing of predictive models consist...
null
null
liu21n_interspeech
A Speech Emotion Recognition Framework for Better Discrimination of Confusions
[ "Jiawang Liu", "Haoxiang Wang" ]
https://www.isca-archive.org/interspeech_2021/liu21n_interspeech.html
https://www.isca-archive.org/interspeech_2021/liu21n_interspeech.pdf
10.21437/Interspeech.2021-718
4483-4487
@inproceedings{liu21n_interspeech, title = {{A Speech Emotion Recognition Framework for Better Discrimination of Confusions}}, author = {Jiawang Liu and Haoxiang Wang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4483--4487}, doi = {10.21437/Interspeech.2021-718}, issn ...
Speech emotion recognition (SER) plays an important role in human-machine interaction (HMI). Various methods have been proposed for the SER task. However, a common problem in most of the previous studies is some specific emotions are grossly misclassified. In this paper, we propose a novel SER framework aiming at discr...
null
null
li21p_interspeech
Speech Emotion Recognition via Multi-Level Cross-Modal Distillation
[ "Ruichen Li", "Jinming Zhao", "Qin Jin" ]
https://www.isca-archive.org/interspeech_2021/li21p_interspeech.html
https://www.isca-archive.org/interspeech_2021/li21p_interspeech.pdf
10.21437/Interspeech.2021-785
4488-4492
@inproceedings{li21p_interspeech, title = {{Speech Emotion Recognition via Multi-Level Cross-Modal Distillation}}, author = {Ruichen Li and Jinming Zhao and Qin Jin}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4488--4492}, doi = {10.21437/Interspeech.2021-785}, issn ...
Speech emotion recognition faces the problem that most of the existing speech corpora are limited in scale and diversity due to the high annotation cost and label ambiguity. In this work, we explore the task of learning robust speech emotion representations based on large unlabeled speech data. Under a simple assumptio...
null
null
ito21_interspeech
Audio-Visual Speech Emotion Recognition by Disentangling Emotion and Identity Attributes
[ "Koichiro Ito", "Takuya Fujioka", "Qinghua Sun", "Kenji Nagamatsu" ]
https://www.isca-archive.org/interspeech_2021/ito21_interspeech.html
https://www.isca-archive.org/interspeech_2021/ito21_interspeech.pdf
10.21437/Interspeech.2021-809
4493-4497
@inproceedings{ito21_interspeech, title = {{Audio-Visual Speech Emotion Recognition by Disentangling Emotion and Identity Attributes}}, author = {Koichiro Ito and Takuya Fujioka and Qinghua Sun and Kenji Nagamatsu}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4493--4497}, doi ...
In this paper, we propose an audio-visual speech emotion recognition (AV-SER) that can suppress the disturbance from an identity attribute by disentangling an emotion attribute and an identity one. We developed a model that first disentangles both attributes for each modality. In order to achieve the disentanglement, w...
null
null
bose21_interspeech
Parametric Distributions to Model Numerical Emotion Labels
[ "Deboshree Bose", "Vidhyasaharan Sethu", "Eliathamby Ambikairajah" ]
https://www.isca-archive.org/interspeech_2021/bose21_interspeech.html
https://www.isca-archive.org/interspeech_2021/bose21_interspeech.pdf
10.21437/Interspeech.2021-1000
4498-4502
@inproceedings{bose21_interspeech, title = {{Parametric Distributions to Model Numerical Emotion Labels}}, author = {Deboshree Bose and Vidhyasaharan Sethu and Eliathamby Ambikairajah}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4498--4502}, doi = {10.21437/Interspeech.2...
It is common to represent emotional states as values on a set of numerical scales corresponding to attributes such as arousal and valence. Often these labels are obtained from multiple annotators who record their perception of emotion in terms of these attributes. Combining these multiple annotations by taking the mean...
null
null
gao21e_interspeech
Metric Learning Based Feature Representation with Gated Fusion Model for Speech Emotion Recognition
[ "Yuan Gao", "Jiaxing Liu", "Longbiao Wang", "Jianwu Dang" ]
https://www.isca-archive.org/interspeech_2021/gao21e_interspeech.html
https://www.isca-archive.org/interspeech_2021/gao21e_interspeech.pdf
10.21437/Interspeech.2021-1133
4503-4507
@inproceedings{gao21e_interspeech, title = {{Metric Learning Based Feature Representation with Gated Fusion Model for Speech Emotion Recognition}}, author = {Yuan Gao and Jiaxing Liu and Longbiao Wang and Jianwu Dang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4503--4507}, do...
Due to the lack of sufficient speech emotional data, the recognition performance of existing speech emotion recognition (SER) approaches is relatively low and requires further improvement to meet the needs of real-life applications. For the problem of data scarcity, an increasingly popular solution is to transfer emoti...
null
null
cai21b_interspeech
Speech Emotion Recognition with Multi-Task Learning
[ "Xingyu Cai", "Jiahong Yuan", "Renjie Zheng", "Liang Huang", "Kenneth Church" ]
https://www.isca-archive.org/interspeech_2021/cai21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/cai21b_interspeech.pdf
10.21437/Interspeech.2021-1852
4508-4512
@inproceedings{cai21b_interspeech, title = {{Speech Emotion Recognition with Multi-Task Learning}}, author = {Xingyu Cai and Jiahong Yuan and Renjie Zheng and Liang Huang and Kenneth Church}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4508--4512}, doi = {10.21437/Intersp...
Speech emotion recognition (SER) classifies speech into emotion categories such as: Happy, Angry, Sad and Neutral . Recently, deep learning has been applied to the SER task. This paper proposes a multi-task learning (MTL) framework to simultaneously perform speech-to-text recognition and emotion classification, with an...
null
null
seneviratne21b_interspeech
Generalized Dilated CNN Models for Depression Detection Using Inverted Vocal Tract Variables
[ "Nadee Seneviratne", "Carol Espy-Wilson" ]
https://www.isca-archive.org/interspeech_2021/seneviratne21b_interspeech.html
https://www.isca-archive.org/interspeech_2021/seneviratne21b_interspeech.pdf
10.21437/Interspeech.2021-1960
4513-4517
@inproceedings{seneviratne21b_interspeech, title = {{Generalized Dilated CNN Models for Depression Detection Using Inverted Vocal Tract Variables}}, author = {Nadee Seneviratne and Carol Espy-Wilson}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4513--4517}, doi = {10.2143...
Depression detection using vocal biomarkers is a highly researched area. Articulatory coordination features (ACFs) are developed based on the changes in neuromotor coordination due to psychomotor slowing, a key feature of Major Depressive Disorder. However findings of existing studies are mostly validated on a single d...
2011.06739
title_snapshot
wang21ga_interspeech
Learning Mutual Correlation in Multimodal Transformer for Speech Emotion Recognition
[ "Yuhua Wang", "Guang Shen", "Yuezhu Xu", "Jiahang Li", "Zhengdao Zhao" ]
https://www.isca-archive.org/interspeech_2021/wang21ga_interspeech.html
https://www.isca-archive.org/interspeech_2021/wang21ga_interspeech.pdf
10.21437/Interspeech.2021-2004
4518-4522
@inproceedings{wang21ga_interspeech, title = {{Learning Mutual Correlation in Multimodal Transformer for Speech Emotion Recognition}}, author = {Yuhua Wang and Guang Shen and Yuezhu Xu and Jiahang Li and Zhengdao Zhao}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4518--4522}, d...
Various studies have confirmed the necessity and benefits of leveraging multimodal features for SER, and the latest research results show that the temporal information captured by the transformer is very useful for improving multimodal speech emotion recognition. However, the dependency between different modalities and...
null
null
liu21o_interspeech
Time-Frequency Representation Learning with Graph Convolutional Network for Dialogue-Level Speech Emotion Recognition
[ "Jiaxing Liu", "Yaodong Song", "Longbiao Wang", "Jianwu Dang", "Ruiguo Yu" ]
https://www.isca-archive.org/interspeech_2021/liu21o_interspeech.html
https://www.isca-archive.org/interspeech_2021/liu21o_interspeech.pdf
10.21437/Interspeech.2021-2067
4523-4527
@inproceedings{liu21o_interspeech, title = {{Time-Frequency Representation Learning with Graph Convolutional Network for Dialogue-Level Speech Emotion Recognition}}, author = {Jiaxing Liu and Yaodong Song and Longbiao Wang and Jianwu Dang and Ruiguo Yu}, year = {2021}, booktitle = {{Interspeech 2021...
With the development of speech emotion recognition (SER), dialogue-level SER (DSER) is more aligned with actual scenarios. In this paper, we propose a DSER approach that includes two stages of representation learning: intra-utterance representation learning and inter-utterance representation learning. In the intra-utte...
null
null
mordido21_interspeech
Compressing 1D Time-Channel Separable Convolutions Using Sparse Random Ternary Matrices
[ "Gonçalo Mordido", "Matthijs Van keirsbilck", "Alexander Keller" ]
https://www.isca-archive.org/interspeech_2021/mordido21_interspeech.html
https://www.isca-archive.org/interspeech_2021/mordido21_interspeech.pdf
10.21437/Interspeech.2021-141
4528-4532
@inproceedings{mordido21_interspeech, title = {{Compressing 1D Time-Channel Separable Convolutions Using Sparse Random Ternary Matrices}}, author = {Gonçalo Mordido and Matthijs {Van keirsbilck} and Alexander Keller}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4528--4532}, doi...
We demonstrate that 1×1-convolutions in 1D time-channel separable convolutions may be replaced by constant, sparse random ternary matrices with weights in -1, 0, +1. Such layers do not perform any multiplications and do not require training. Moreover, the matrices may be generated on the chip during computation and the...
2103.17142
title_snapshot
cheng21c_interspeech
Weakly Supervised Construction of ASR Systems from Massive Video Data
[ "Mengli Cheng", "Chengyu Wang", "Jun Huang", "Xiaobo Wang" ]
https://www.isca-archive.org/interspeech_2021/cheng21c_interspeech.html
https://www.isca-archive.org/interspeech_2021/cheng21c_interspeech.pdf
10.21437/Interspeech.2021-7
4533-4537
@inproceedings{cheng21c_interspeech, title = {{Weakly Supervised Construction of ASR Systems from Massive Video Data}}, author = {Mengli Cheng and Chengyu Wang and Jun Huang and Xiaobo Wang}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4533--4537}, doi = {10.21437/Intersp...
Despite the rapid development of deep learning models, for real-world applications, building large-scale Automatic Speech Recognition (ASR) systems from scratch is still significantly challenging, mostly due to the time-consuming and financially-expensive process of annotating a large amount of audio data with transcri...
2008.01300
title_judge
kim21l_interspeech
Broadcasted Residual Learning for Efficient Keyword Spotting
[ "Byeonggeun Kim", "Simyung Chang", "Jinkyu Lee", "Dooyong Sung" ]
https://www.isca-archive.org/interspeech_2021/kim21l_interspeech.html
https://www.isca-archive.org/interspeech_2021/kim21l_interspeech.pdf
10.21437/Interspeech.2021-383
4538-4542
@inproceedings{kim21l_interspeech, title = {{Broadcasted Residual Learning for Efficient Keyword Spotting}}, author = {Byeonggeun Kim and Simyung Chang and Jinkyu Lee and Dooyong Sung}, year = {2021}, booktitle = {{Interspeech 2021}}, pages = {4538--4542}, doi = {10.21437/Interspeech.2...
Keyword spotting is an important research field because it plays a key role in device wake-up and user interaction on smart devices. However, it is challenging to minimize errors while operating efficiently in devices with limited resources such as mobile phones. We present a broadcasted residual learning method to ach...
2106.04140
title_snapshot