An Intelligent System for Spoken Term Detection That Uses Belief Combination
- Submitting institution
-
Liverpool John Moores University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 938
- Type
- D - Journal article
- DOI
-
10.1109/mis.2017.13
- Title of journal
- IEEE Intelligent Systems
- Article number
- -
- First page
- 70
- Volume
- 32
- Issue
- 1
- ISSN
- 1541-1672
- Open access status
- Compliant
- Month of publication
- February
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
1
- Research group(s)
-
-
- Citation count
- 6
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The underpinning research introduces a hybrid decision model based on composite algorithms, to handle the challenges related to template matching of continuous speech signals due to the dynamic nature of speech. For the first time, the study utilises the ‘theory of evidence’ in speech analysis to handle the uncertainty in beliefs from multiple sources. Standard experimental design using large speech datasets showed that the model achieved outstanding performance compared to existing speaker-dependent spoken-term-detection methods. The system has many uses in relation to unsupervised speech analysis, such as YouTube content matching and query-search in audio class.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
