In this paper, we have integrated in a GMM based speaker identi cation system two di erent techniques: a) Maximum Likelihood Linear Regression (MLLR) transformation which adapts the system to the new environment based on modifying the continuous densities of the GMM mixtures. We apply the MLLR to perform environmental compensation by reducing a mismatch due to channel or additive noise e ects, ...

In this paper, the theoretical framework of maximum a posterior linear regression (MAPLR) based variance adaptation for continuous density HMMs is described. In our approach, a class of informative prior distribution for MAPLR based variance adaptation is identified, from which the close form solution of MAPLR based variance adaptation is obtained under its EM formulation. Effects of the propos...

In this paper, we investigate automatic language proficiency assessment from learners’ utterances generated through shadowing and reading aloud. By increasing the degrees of difficulty of learners’ tasks for each practice, we examine how the automatic scores, the conventional GOP and proposed F-GOP, change according to the cognitive loads posed on learners. We also investigate the effect and si...

We propose a technique for generating a large amount of target speaker-like speech features by converting a large amount of prepared speech features of many speakers into features similar to those of the target speaker using a transformation matrix. To generate a large amount of target speaker-like features, the system only needs a very small amount of the target speaker’s utterances. This tech...

In this paper, we present several adaptation methods for nonnative speech recognition. We have tested pronunciation modelling, MLLR and MAP non-native pronunciation adaptation and HMM models retraining on the HIWIRE foreign accented English speech database. The “phonetic confusion” scheme we have developed consists in associating to each spoken phone several sequences of confused phones. In our...

This paper describes a technique for synthesizing speech with any desired voice. The technique is based on an HMM-based text-to-speech (TTS) system and MLLR adaptation algorithm. To generate speech of an arbitrarily given target speaker, speaker-independent speech units, i.e., average voice models, is adapted to the target speaker using MLLR framework. In addition to spectrum and pitch adaptati...

Recently, we have been investigating the application of kernel methods to improve the performance of eigenvoice-based adaptation methods by exploiting possible nonlinearity in their original working space. We proposed the kernel eigenvoice adaptation (KEV) in [1], and the kernel eigenspace-based MLLR adaptation (KEMLLR) in [2]. In KEMLLR, speaker-dependent MLLR transformation matrices are mappe...

We have been investigating the use of kernel methods to improve conventional linear adaptation algorithms for fast adaptation, when there are less than 10s of adaptation speech. On clean speech, we had shown that our new kernel-based adaptation methods, namely, embedded kernel eigenvoice (eKEV) and kernel eigenspace-based MLLR (KEMLLR) outperformed their linear counterparts. In this paper, we s...

This thesis considers the entire automated speech recognition process and presents a standardised approach to LVCSR experimentation with HMMs. It also discusses various approaches to speaker adaptation such as MLLR and multiscale, and presents experimental results for cross-task speaker adaptation. An analysis of training parameters and data su ciency for reasonable system performance estimates...

In this paper, we investigate the noise robustness properties of frame-based and sparse point process-based models for spotting keywords in continuous speech. We introduce a new strategy to improve point process model (PPM) robustness by adapting low-level feature detector thresholds to preserve background firing rates in the presence of noise. We find that this unsupervised approach can signif...