The next seminar in the series Linguistics Circle Seminar Series will be held on Friday 13 May at 12:00 in Room 256, M.A. Vassalli Conference Centre - Gateway Building (GW256). During the seminar Andrea De Marco, University of Malta, will talk about 'Acoustic Approaches to Accent Identification'.
Abstract
There has been considerable research on the problems of speaker and language recognition from samples of speech. A less researched problem is that of accent recognition. Although this is a similar problem to language identification, different accents of a language exhibit more fine-grained differences between classes than languages. This presents a tougher problem for traditional classification techniques. This talk will go over recent work and evaluate a number of techniques for accent classification. The proposed techniques are novel modifications and extensions to state of the art algorithms, and they result in enhanced performance on accent recognition.
The bulk of the work is concerned with the application of the i-Vector technique to accent identification, which is the most successful approach to acoustic classification to have emerged in recent years. We show that it is possible to achieve high accuracy accent identification without reliance on transcriptions and without utilising phoneme recognition algorithms. The seminar will describe the various stages in the development of i-Vector based accent classification that improve the standard approaches usually applied for speaker or language identification, which are insufficient. We demonstrate that very good accent identification performance is possible with acoustic methods by considering different i-Vector projections, frontend parameters, i-Vector configuration parameters, and an optimised fusion of the resulting i-Vector classifiers we can obtain from the same data.
The overall claim is that of having achieved the best accent identification performance on the test corpus for acoustic methods, with up to 90\% identification rate. This performance is even better than previously reported acoustic-phonotactic based systems on the same corpus, and is very close to performance obtained via transcription based accent identification. The seminar will also go over the utilization of this technique for speech recognition purposes, leading to considerably lower word error rates.
