2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2017
DOI: 10.1109/icassp.2017.7953122
|View full text |Cite
|
Sign up to set email alerts
|

Automatic assessment of dysarthria severity level using audio descriptors

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
17
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
6
3
1

Relationship

1
9

Authors

Journals

citations
Cited by 46 publications
(17 citation statements)
references
References 11 publications
0
17
0
Order By: Relevance
“…Furthermore, isolated words are also considered in their dataset, whereas we only considered continuous speech. Similarly, the study in Reference [63] uses MFCCs along with delta features in a CNN classifier with the TORGO dataset and reaches a severity assessment accuracy of 98.3%. However, this study also excludes healthy speech data and only conducts testing on two speakers (one mild and one severe), whereas the current study examines all speakers for testing.…”
Section: Discussionmentioning
confidence: 99%
“…Furthermore, isolated words are also considered in their dataset, whereas we only considered continuous speech. Similarly, the study in Reference [63] uses MFCCs along with delta features in a CNN classifier with the TORGO dataset and reaches a severity assessment accuracy of 98.3%. However, this study also excludes healthy speech data and only conducts testing on two speakers (one mild and one severe), whereas the current study examines all speakers for testing.…”
Section: Discussionmentioning
confidence: 99%
“…The advancement in signal processing and ML literature has encouraged researchers to explore the use of signal processing and ML techniques for automatic speech intelligibility assessment. Broadly, these methods aim at measuring the abnormalities in spoken speech by extracting handcrafted acoustic features based on statistical signal processing [14,15] and/or supervised methods based on ML [16,17]. Such techniques offer the advantage of frequent, cost effective and objective assessment of speech intelligibility.…”
Section: Review Of Intelligibility Assessment Methodsmentioning
confidence: 99%
“…This relatively small number is in contrast to the number of speakers in other areas of speech technology, such as speech recognition [72] and speaker verification [73], where it is possible to collect speech from healthy speakers by recording utterances from hundreds or even thousands of talkers. It should be noted, however, that in dysarthric speech assessment, a similar small number of speakers (less than 15) has been used also in previous studies [23][74] [75].…”
Section: The Ua-speech Databasementioning
confidence: 99%