Sorry, you need to enable JavaScript to visit this website.

Audio and Acoustic Signal Processing

BIOLOGICALLY INSPIRED SPEECH EMOTION RECOGNITION


Conventional feature-based classification methods do not apply well to automatic recognition of speech emotions, mostly because the precise set of spectral and prosodic features that is required to identify the emotional state of a speaker has not been determined yet. This paper presents a method that operates directly on the speech signal, thus avoiding the problematic step of feature extraction.

Paper Details

Authors:
Submitted On:
16 March 2017 - 10:05am
Short Link:
Type:
Event:
Document Year:
Cite

Document Files

ICASSP2017_Lotfidereshgi (poster) V2.pdf

(28 downloads)

Keywords

Additional Categories

Subscribe

[1] , "BIOLOGICALLY INSPIRED SPEECH EMOTION RECOGNITION", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1516. Accessed: May. 25, 2017.
@article{1516-17,
url = {http://sigport.org/1516},
author = { },
publisher = {IEEE SigPort},
title = {BIOLOGICALLY INSPIRED SPEECH EMOTION RECOGNITION},
year = {2017} }
TY - EJOUR
T1 - BIOLOGICALLY INSPIRED SPEECH EMOTION RECOGNITION
AU -
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1516
ER -
. (2017). BIOLOGICALLY INSPIRED SPEECH EMOTION RECOGNITION. IEEE SigPort. http://sigport.org/1516
, 2017. BIOLOGICALLY INSPIRED SPEECH EMOTION RECOGNITION. Available at: http://sigport.org/1516.
. (2017). "BIOLOGICALLY INSPIRED SPEECH EMOTION RECOGNITION." Web.
1. . BIOLOGICALLY INSPIRED SPEECH EMOTION RECOGNITION [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1516

ROBUST AUTOMATIC RECOGNITION OF SPEECH WITH BACKGROUND MUSIC


This paper addresses the task of Automatic Speech Recognition (ASR) with music in the background, where the accuracy of recognition may deteriorate significantly.
To improve the robustness of ASR in this task, e.g. for broadcast news transcription or subtitles creation, we adopt two approaches:
1) multi-condition training of the acoustic models and 2) denoising autoencoders followed by acoustic model training on the preprocessed data.
In the latter case, two types of autoencoders are considered: the fully connected and the convolutional network.

Paper Details

Authors:
Jiri Malek, Jindrich Zdansky, Petr Cerva
Submitted On:
28 February 2017 - 9:22am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

posterICASSP2017_MalekZdanskyCerva.pdf

(40 downloads)

Keywords

Additional Categories

Subscribe

[1] Jiri Malek, Jindrich Zdansky, Petr Cerva, "ROBUST AUTOMATIC RECOGNITION OF SPEECH WITH BACKGROUND MUSIC", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1511. Accessed: May. 25, 2017.
@article{1511-17,
url = {http://sigport.org/1511},
author = {Jiri Malek; Jindrich Zdansky; Petr Cerva },
publisher = {IEEE SigPort},
title = {ROBUST AUTOMATIC RECOGNITION OF SPEECH WITH BACKGROUND MUSIC},
year = {2017} }
TY - EJOUR
T1 - ROBUST AUTOMATIC RECOGNITION OF SPEECH WITH BACKGROUND MUSIC
AU - Jiri Malek; Jindrich Zdansky; Petr Cerva
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1511
ER -
Jiri Malek, Jindrich Zdansky, Petr Cerva. (2017). ROBUST AUTOMATIC RECOGNITION OF SPEECH WITH BACKGROUND MUSIC. IEEE SigPort. http://sigport.org/1511
Jiri Malek, Jindrich Zdansky, Petr Cerva, 2017. ROBUST AUTOMATIC RECOGNITION OF SPEECH WITH BACKGROUND MUSIC. Available at: http://sigport.org/1511.
Jiri Malek, Jindrich Zdansky, Petr Cerva. (2017). "ROBUST AUTOMATIC RECOGNITION OF SPEECH WITH BACKGROUND MUSIC." Web.
1. Jiri Malek, Jindrich Zdansky, Petr Cerva. ROBUST AUTOMATIC RECOGNITION OF SPEECH WITH BACKGROUND MUSIC [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1511

Deductive Refinement of Species Labelling in Weakly Labelled Birdsong Recordings


Many approaches have been used in bird species classification from their sound in order to provide labels for the whole of a recording. However, a more precise classification of each bird vocalization would be of great importance to the use and management of sound archives and bird monitoring. In this work, we introduce a technique that using a two step process can first automatically detect all bird vocalizations and then, with the use of ‘weakly’ labelled recordings, classify them.

main.pdf

PDF icon main.pdf (181 downloads)

Paper Details

Authors:
Veronica Morfi, Dan Stowell
Submitted On:
28 February 2017 - 7:05am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

main.pdf

(181 downloads)

Keywords

Subscribe

[1] Veronica Morfi, Dan Stowell, "Deductive Refinement of Species Labelling in Weakly Labelled Birdsong Recordings", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1505. Accessed: May. 25, 2017.
@article{1505-17,
url = {http://sigport.org/1505},
author = {Veronica Morfi; Dan Stowell },
publisher = {IEEE SigPort},
title = {Deductive Refinement of Species Labelling in Weakly Labelled Birdsong Recordings},
year = {2017} }
TY - EJOUR
T1 - Deductive Refinement of Species Labelling in Weakly Labelled Birdsong Recordings
AU - Veronica Morfi; Dan Stowell
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1505
ER -
Veronica Morfi, Dan Stowell. (2017). Deductive Refinement of Species Labelling in Weakly Labelled Birdsong Recordings. IEEE SigPort. http://sigport.org/1505
Veronica Morfi, Dan Stowell, 2017. Deductive Refinement of Species Labelling in Weakly Labelled Birdsong Recordings. Available at: http://sigport.org/1505.
Veronica Morfi, Dan Stowell. (2017). "Deductive Refinement of Species Labelling in Weakly Labelled Birdsong Recordings." Web.
1. Veronica Morfi, Dan Stowell. Deductive Refinement of Species Labelling in Weakly Labelled Birdsong Recordings [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1505

Application of Compressed Sensing to Wideband Spectrum Sensing in Cognitive Radio Networks

Paper Details

Authors:
Michael M. Abdel-Sayed, Ahmed Khattab, Mohamed F. Abu-Elyazeed
Submitted On:
28 February 2017 - 2:59am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP_Poster_4338.pdf

(33 downloads)

Keywords

Additional Categories

Subscribe

[1] Michael M. Abdel-Sayed, Ahmed Khattab, Mohamed F. Abu-Elyazeed, "Application of Compressed Sensing to Wideband Spectrum Sensing in Cognitive Radio Networks", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1487. Accessed: May. 25, 2017.
@article{1487-17,
url = {http://sigport.org/1487},
author = {Michael M. Abdel-Sayed; Ahmed Khattab; Mohamed F. Abu-Elyazeed },
publisher = {IEEE SigPort},
title = {Application of Compressed Sensing to Wideband Spectrum Sensing in Cognitive Radio Networks},
year = {2017} }
TY - EJOUR
T1 - Application of Compressed Sensing to Wideband Spectrum Sensing in Cognitive Radio Networks
AU - Michael M. Abdel-Sayed; Ahmed Khattab; Mohamed F. Abu-Elyazeed
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1487
ER -
Michael M. Abdel-Sayed, Ahmed Khattab, Mohamed F. Abu-Elyazeed. (2017). Application of Compressed Sensing to Wideband Spectrum Sensing in Cognitive Radio Networks. IEEE SigPort. http://sigport.org/1487
Michael M. Abdel-Sayed, Ahmed Khattab, Mohamed F. Abu-Elyazeed, 2017. Application of Compressed Sensing to Wideband Spectrum Sensing in Cognitive Radio Networks. Available at: http://sigport.org/1487.
Michael M. Abdel-Sayed, Ahmed Khattab, Mohamed F. Abu-Elyazeed. (2017). "Application of Compressed Sensing to Wideband Spectrum Sensing in Cognitive Radio Networks." Web.
1. Michael M. Abdel-Sayed, Ahmed Khattab, Mohamed F. Abu-Elyazeed. Application of Compressed Sensing to Wideband Spectrum Sensing in Cognitive Radio Networks [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1487

FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS


Speaker diarization in noisy conditions is addressed in this paper. The regression-based DNN is first adopted to map the noisy acoustic features to the clean features, and then consensus clustering of the original and mapped features is used to fuse the diarization results. The experiments are conducted on the IFLY-DIAR-II database, which is a Chinese talk show database with various noise types, such as music, applause and laughter. Compared to the baseline system using PLP features, a 21.26% relative DER improvement can be achieved using the proposed algorithm.

Paper Details

Authors:
Weixin Zhu, Wu Guo, Guoping Hu
Submitted On:
11 March 2017 - 8:49pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

poster_wxzhu_v2.pptx

(50 downloads)

Keywords

Subscribe

[1] Weixin Zhu, Wu Guo, Guoping Hu, "FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1460. Accessed: May. 25, 2017.
@article{1460-17,
url = {http://sigport.org/1460},
author = {Weixin Zhu; Wu Guo; Guoping Hu },
publisher = {IEEE SigPort},
title = {FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS},
year = {2017} }
TY - EJOUR
T1 - FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS
AU - Weixin Zhu; Wu Guo; Guoping Hu
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1460
ER -
Weixin Zhu, Wu Guo, Guoping Hu. (2017). FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS. IEEE SigPort. http://sigport.org/1460
Weixin Zhu, Wu Guo, Guoping Hu, 2017. FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS. Available at: http://sigport.org/1460.
Weixin Zhu, Wu Guo, Guoping Hu. (2017). "FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS." Web.
1. Weixin Zhu, Wu Guo, Guoping Hu. FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1460

FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS


Speaker diarization in noisy conditions is addressed in this paper. The regression-based DNN is first adopted to map the noisy acoustic features to the clean features, and then consensus clustering of the original and mapped features is used to fuse the diarization results. The experiments are conducted on the IFLY-DIAR-II database, which is a Chinese talk show database with various noise types, such as music, applause and laughter. Compared to the baseline system using PLP features, a 21.26% relative DER improvement can be achieved using the proposed algorithm.

Paper Details

Authors:
Weixin Zhu, Wu Guo, Guoping Hu
Submitted On:
27 February 2017 - 8:37pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

poster_wxzhu_v2.pptx

(50 downloads)

Keywords

Subscribe

[1] Weixin Zhu, Wu Guo, Guoping Hu, "FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1459. Accessed: May. 25, 2017.
@article{1459-17,
url = {http://sigport.org/1459},
author = {Weixin Zhu; Wu Guo; Guoping Hu },
publisher = {IEEE SigPort},
title = {FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS},
year = {2017} }
TY - EJOUR
T1 - FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS
AU - Weixin Zhu; Wu Guo; Guoping Hu
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1459
ER -
Weixin Zhu, Wu Guo, Guoping Hu. (2017). FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS. IEEE SigPort. http://sigport.org/1459
Weixin Zhu, Wu Guo, Guoping Hu, 2017. FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS. Available at: http://sigport.org/1459.
Weixin Zhu, Wu Guo, Guoping Hu. (2017). "FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS." Web.
1. Weixin Zhu, Wu Guo, Guoping Hu. FEATURE MAPPING FOR SPEAKER DIARIZATION IN NOISY CONDITIONS [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1459

Discovering Sound Concepts and Acoustic Relations in Text


In this paper we describe approaches for discovering acoustic concepts and relations in text. The first major goal is to be able to identify text phrases which contain a notion of audibility and can be termed as a sound or an acoustic concept. We also propose a method to define an acoustic scene through a set of sound concepts. We use pattern matching and parts of speech tags to generate sound concepts from large scale text corpora. We use dependency parsing

Paper Details

Authors:
Bhiksha Raj, Ndapandula Nakashole
Submitted On:
27 February 2017 - 8:22pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

anurag_icassp17.pdf

(31 downloads)

Keywords

Subscribe

[1] Bhiksha Raj, Ndapandula Nakashole, "Discovering Sound Concepts and Acoustic Relations in Text", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1455. Accessed: May. 25, 2017.
@article{1455-17,
url = {http://sigport.org/1455},
author = {Bhiksha Raj; Ndapandula Nakashole },
publisher = {IEEE SigPort},
title = {Discovering Sound Concepts and Acoustic Relations in Text},
year = {2017} }
TY - EJOUR
T1 - Discovering Sound Concepts and Acoustic Relations in Text
AU - Bhiksha Raj; Ndapandula Nakashole
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1455
ER -
Bhiksha Raj, Ndapandula Nakashole. (2017). Discovering Sound Concepts and Acoustic Relations in Text. IEEE SigPort. http://sigport.org/1455
Bhiksha Raj, Ndapandula Nakashole, 2017. Discovering Sound Concepts and Acoustic Relations in Text. Available at: http://sigport.org/1455.
Bhiksha Raj, Ndapandula Nakashole. (2017). "Discovering Sound Concepts and Acoustic Relations in Text." Web.
1. Bhiksha Raj, Ndapandula Nakashole. Discovering Sound Concepts and Acoustic Relations in Text [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1455

First Investigation of Universal Speech Attributes for Speaker Verification

Paper Details

Authors:
Submitted On:
27 February 2017 - 8:54pm
Short Link:
Type:
Event:

Document Files

First Investigation of Universal Speech Attributes for Speaker Verification.pdf

(63 downloads)

Keywords

Subscribe

[1] , "First Investigation of Universal Speech Attributes for Speaker Verification", IEEE SigPort, 2016. [Online]. Available: http://sigport.org/1440. Accessed: May. 25, 2017.
@article{1440-16,
url = {http://sigport.org/1440},
author = { },
publisher = {IEEE SigPort},
title = {First Investigation of Universal Speech Attributes for Speaker Verification},
year = {2016} }
TY - EJOUR
T1 - First Investigation of Universal Speech Attributes for Speaker Verification
AU -
PY - 2016
PB - IEEE SigPort
UR - http://sigport.org/1440
ER -
. (2016). First Investigation of Universal Speech Attributes for Speaker Verification. IEEE SigPort. http://sigport.org/1440
, 2016. First Investigation of Universal Speech Attributes for Speaker Verification. Available at: http://sigport.org/1440.
. (2016). "First Investigation of Universal Speech Attributes for Speaker Verification." Web.
1. . First Investigation of Universal Speech Attributes for Speaker Verification [Internet]. IEEE SigPort; 2016. Available from : http://sigport.org/1440

Graph-Based Active Learning: A New Look at Expected Error Minimization


In graph-based active learning, algorithms based on expected error minimization (EEM) have been popular and yield good empirical performance.
The exact computation of EEM optimally balances exploration and exploitation.
In practice, however, EEM-based algorithms employ various approximations due to the computational hardness of exact EEM.
This can result in a lack of either exploration or exploitation, which can negatively impact the effectiveness of active learning.

Paper Details

Authors:
Kwang-Sung Jun, Robert Nowak
Submitted On:
8 December 2016 - 4:48pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

graphal-globalsip-1208.pdf

(58 downloads)

Keywords

Additional Categories

Subscribe

[1] Kwang-Sung Jun, Robert Nowak, "Graph-Based Active Learning: A New Look at Expected Error Minimization", IEEE SigPort, 2016. [Online]. Available: http://sigport.org/1427. Accessed: May. 25, 2017.
@article{1427-16,
url = {http://sigport.org/1427},
author = {Kwang-Sung Jun; Robert Nowak },
publisher = {IEEE SigPort},
title = {Graph-Based Active Learning: A New Look at Expected Error Minimization},
year = {2016} }
TY - EJOUR
T1 - Graph-Based Active Learning: A New Look at Expected Error Minimization
AU - Kwang-Sung Jun; Robert Nowak
PY - 2016
PB - IEEE SigPort
UR - http://sigport.org/1427
ER -
Kwang-Sung Jun, Robert Nowak. (2016). Graph-Based Active Learning: A New Look at Expected Error Minimization. IEEE SigPort. http://sigport.org/1427
Kwang-Sung Jun, Robert Nowak, 2016. Graph-Based Active Learning: A New Look at Expected Error Minimization. Available at: http://sigport.org/1427.
Kwang-Sung Jun, Robert Nowak. (2016). "Graph-Based Active Learning: A New Look at Expected Error Minimization." Web.
1. Kwang-Sung Jun, Robert Nowak. Graph-Based Active Learning: A New Look at Expected Error Minimization [Internet]. IEEE SigPort; 2016. Available from : http://sigport.org/1427

Cyber-Resilient Control of Inverter Based Microgrids


This paper investigates cyber-attacks compromising the data integrity of 100% inverter-interfaced islanded microgrids featuring an energy storage system (ESS), a wind turbine generator (WTG), a photovoltaic system (PV) and controllable loads. The ESS operates as the isochronous generator, responsible for forming the microgrid voltage and frequency, whereas the WTG and PV distributed energy resources (DER) operate in maximum-power-point-tracking (MPPT) mode.

Paper Details

Authors:
Martine Chlela, Diego Mascarella, Geza Joos, Marthe Kassouf
Submitted On:
7 December 2016 - 3:56am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

GlobalSIP_MartineChlela_8Dec2016.pdf

(55 downloads)

Keywords

Additional Categories

Subscribe

[1] Martine Chlela, Diego Mascarella, Geza Joos, Marthe Kassouf, "Cyber-Resilient Control of Inverter Based Microgrids", IEEE SigPort, 2016. [Online]. Available: http://sigport.org/1402. Accessed: May. 25, 2017.
@article{1402-16,
url = {http://sigport.org/1402},
author = {Martine Chlela; Diego Mascarella; Geza Joos; Marthe Kassouf },
publisher = {IEEE SigPort},
title = {Cyber-Resilient Control of Inverter Based Microgrids},
year = {2016} }
TY - EJOUR
T1 - Cyber-Resilient Control of Inverter Based Microgrids
AU - Martine Chlela; Diego Mascarella; Geza Joos; Marthe Kassouf
PY - 2016
PB - IEEE SigPort
UR - http://sigport.org/1402
ER -
Martine Chlela, Diego Mascarella, Geza Joos, Marthe Kassouf. (2016). Cyber-Resilient Control of Inverter Based Microgrids. IEEE SigPort. http://sigport.org/1402
Martine Chlela, Diego Mascarella, Geza Joos, Marthe Kassouf, 2016. Cyber-Resilient Control of Inverter Based Microgrids. Available at: http://sigport.org/1402.
Martine Chlela, Diego Mascarella, Geza Joos, Marthe Kassouf. (2016). "Cyber-Resilient Control of Inverter Based Microgrids." Web.
1. Martine Chlela, Diego Mascarella, Geza Joos, Marthe Kassouf. Cyber-Resilient Control of Inverter Based Microgrids [Internet]. IEEE SigPort; 2016. Available from : http://sigport.org/1402

Pages