2013 18th International Conference on Digital Signal Processing (DSP) 2013
DOI: 10.1109/icdsp.2013.6622707
|View full text |Cite
|
Sign up to set email alerts
|

Experiments on far-field multichannel speech processing in smart homes

Abstract: Abstract-In this paper, we examine three problems that rise in the modern, challenging area of far-field speech processing. The developed methods for each problem, namely (a) multichannel speech enhancement, (b) voice activity detection, and (c) speech recognition, are potentially applicable to a distant speech recognition system for voice-enabled smart home environments. The obtained results on real and simulated data, regarding the smart home speech applications, are quite promising due to the accomplished i… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
2
0
2

Year Published

2014
2014
2019
2019

Publication Types

Select...
4
1

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(4 citation statements)
references
References 20 publications
0
2
0
2
Order By: Relevance
“…Table 2 reports the SVM-based classification performance (in terms of F-score) of different feature sets for each room of the DIRHA smart home, using ground-truth room-independent SAD boundaries. In all single feature cases, 5-dimensional vectors are produced (one feature per room), yielding 15-dimensional vectors when all 3 feature sets are considered (see also (9)). Clearly, the combined feature vector outperforms any single feature set, suggesting that different features carry complementary information for room selection.…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…Table 2 reports the SVM-based classification performance (in terms of F-score) of different feature sets for each room of the DIRHA smart home, using ground-truth room-independent SAD boundaries. In all single feature cases, 5-dimensional vectors are produced (one feature per room), yielding 15-dimensional vectors when all 3 feature sets are considered (see also (9)). Clearly, the combined feature vector outperforms any single feature set, suggesting that different features carry complementary information for room selection.…”
Section: Resultsmentioning
confidence: 99%
“…Additionally, ASR can benefit from optimal channel selection among the microphoneequipped rooms [4]. Indeed, as already pointed out in [5][6][7][8][9], various approaches can be successfully applied on multi-channel data, since spatial processing can effectively exploit the information available about the desired sources. Further, results in [4,10] demonstrate the challenges presented to far-field ASR in multi-room environments, specifically showing the crucial impact of SAD in a multimicrophone spoken dialogue system.…”
Section: Introductionmentioning
confidence: 98%
“…Ïpou M e-nai o arijmÏc twn DOA gramm∏n. Praktikà katal †goume se mia l'sh kleist †c morf †c qrhsimopoi∏ntac elàqista tetràgwna (Least Squares (LS)) [208].…”
Section: Ekt-mhsh Jëshc Omilht † Me Elàqista Tetràgwnaunclassified
“…Συνοψίζουμε τα κίνητρα που μας ώθησαν στις ερευνητικές κατευθύνσεις της παρούσας έρευνας καθώς και τις κύριες συνεισφορές της [59,60,[111][112][113][114][115][116][117][118]135].…”
Section: κίνητρα και ερευνητικές συνεισφορέςunclassified