Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Measuring, refining and calibrating speaker and language information extracted from speech
Thesis (PhD (Electrical and Electronic Engineering))--University of Stellenbosch, 2010.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Stellenbosch : University of Stellenbosch
2010
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613773833961473 |
|---|---|
| access_status_str | Open Access |
| author | Brummer, Niko |
| author2 | Du Preez, J. A. |
| author_browse | Brummer, Niko Du Preez, J. A. |
| author_facet | Du Preez, J. A. Brummer, Niko |
| author_sort | Brummer, Niko |
| collection | Thesis |
| dc_rights_str_mv | University of Stellenbosch |
| description | Thesis (PhD (Electrical and Electronic Engineering))--University of Stellenbosch, 2010. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/5139 |
| institution | Stellenbosch University (South Africa) |
| last_indexed | 2026-06-10T12:41:28.315Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2010 |
| publishDateRange | 2010 |
| publishDateSort | 2010 |
| publisher | Stellenbosch : University of Stellenbosch |
| publisherStr | Stellenbosch : University of Stellenbosch |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/5139 Measuring, refining and calibrating speaker and language information extracted from speech Brummer, Niko Du Preez, J. A. University of Stellenbosch. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Automatic speaker recognition Automatic spoken language recognition Proper scoring rule Calibration Dissertations -- Electronic engineering Theses -- Electronic engineering Automatic speech recognition Speech processing systems Thesis (PhD (Electrical and Electronic Engineering))--University of Stellenbosch, 2010. ENGLISH ABSTRACT: We propose a new methodology, based on proper scoring rules, for the evaluation of the goodness of pattern recognizers with probabilistic outputs. The recognizers of interest take an input, known to belong to one of a discrete set of classes, and output a calibrated likelihood for each class. This is a generalization of the traditional use of proper scoring rules to evaluate the goodness of probability distributions. A recognizer with outputs in well-calibrated probability distribution form can be applied to make cost-effective Bayes decisions over a range of applications, having di fferent cost functions. A recognizer with likelihood output can additionally be employed for a wide range of prior distributions for the to-be-recognized classes. We use automatic speaker recognition and automatic spoken language recognition as prototypes of this type of pattern recognizer. The traditional evaluation methods in these fields, as represented by the series of NIST Speaker and Language Recognition Evaluations, evaluate hard decisions made by the recognizers. This makes these recognizers cost-and-prior-dependent. The proposed methodology generalizes that of the NIST evaluations, allowing for the evaluation of recognizers which are intended to be usefully applied over a wide range of applications, having variable priors and costs. The proposal includes a family of evaluation criteria, where each member of the family is formed by a proper scoring rule. We emphasize two members of this family: (i) A non-strict scoring rule, directly representing error-rate at a given prior. (ii) The strict logarithmic scoring rule which represents information content, or which equivalently represents summarized error-rate, or expected cost, over a wide range of applications. We further show how to form a family of secondary evaluation criteria, which by contrasting with the primary criteria, form an analysis of the goodness of calibration of the recognizers likelihoods. Finally, we show how to use the logarithmic scoring rule as an objective function for the discriminative training of fusion and calibration of speaker and language recognizers. AFRIKAANSE OPSOMMING: Ons wys hoe om die onsekerheid in die uittree van outomatiese sprekerherkenning- en taalherkenningstelsels voor te stel, te meet, te kalibreer en te optimeer. Dit maak die bestaande tegnologie akkurater, doeltre ender en meer algemeen toepasbaar. Doctoral 2010-11-15T14:41:37Z 2010-12-15T10:14:24Z 2010-11-15T14:41:37Z 2010-12-15T10:14:24Z 2010-12 Thesis http://hdl.handle.net/10019.1/5139 University of Stellenbosch 160 p. : ill. application/pdf Stellenbosch : University of Stellenbosch |
| spellingShingle | Automatic speaker recognition Automatic spoken language recognition Proper scoring rule Calibration Dissertations -- Electronic engineering Theses -- Electronic engineering Automatic speech recognition Speech processing systems Brummer, Niko Measuring, refining and calibrating speaker and language information extracted from speech |
| title | Measuring, refining and calibrating speaker and language information extracted from speech |
| title_full | Measuring, refining and calibrating speaker and language information extracted from speech |
| title_fullStr | Measuring, refining and calibrating speaker and language information extracted from speech |
| title_full_unstemmed | Measuring, refining and calibrating speaker and language information extracted from speech |
| title_short | Measuring, refining and calibrating speaker and language information extracted from speech |
| title_sort | measuring refining and calibrating speaker and language information extracted from speech |
| topic | Automatic speaker recognition Automatic spoken language recognition Proper scoring rule Calibration Dissertations -- Electronic engineering Theses -- Electronic engineering Automatic speech recognition Speech processing systems |
| url | http://hdl.handle.net/10019.1/5139 |
| work_keys_str_mv | AT brummerniko measuringrefiningandcalibratingspeakerandlanguageinformationextractedfromspeech |