Robust Speaker Recognition in Noisy Conditions

This paper investigates the problem of speaker identi-fication and verification in noisy conditions, assuming that speechsignals are corrupted by environmental noise, but knowledgeabout the noise characteristics is not available. This research ismotivated in part by the potential application of speaker recog-nition technologies on handheld devices or the Internet. Whilethe technologies promise an additional biometric layer of securityto protect the user, the practical implementation of such systemsfaces many challenges. One of these is environmental noise. Due tothe mobile nature of such systems, the noise sources can be highlytime-varying and potentially unknown. This raises the require-ment for noise robustness in the absence of information about thenoise. This paper describes a method that combines multicondi-tion model training and missing-feature theory to model noisewith unknown temporal-spectral characteristics. Multiconditiontraining is conducted using simulated noisy data with limitednoise variation, providing a “coarse” compensation for the noise,and missing-feature theory is applied to refine the compensationby ignoring noise variation outside the given training conditions,thereby reducing the training and testing mismatch. This paperis focused on several issues relating to the implementation of thenew model for real-world applications. These include the gener-ation of multicondition training data to model noisy speech, thecombination of different training data to optimize the recognitionperformance, and the reduction of the model’s complexity. Thenew algorithm was tested using two databases with simulated andrealistic noisy speech data. The first database is a redevelopmentof the TIMIT database by rerecording the data in the presence ofvarious noise types, used to test the model for speaker identifica-tion with a focus on the varieties of noise. The second database isa handheld-device database collected in realistic noisy conditions,used to further validate the model for real-world speaker verifica-tion. The new model is compared to baseline systems and is foundto achieve lower error rates.

Multimodal Biometric Human Recognition for Perceptual Human-Computer Interaction

In this paper, a novel video-based multimodal biometric verification scheme using the subspace-based low-level feature fusion of face and speech is developed for specific speaker recognition for perceptual human--computer interaction (HCI). In the proposed scheme, human face is tracked and face pose is estimated to weight the detected facelike regions in successive frames, where ill-posed faces and false-positive detections are assigned with lower credit to enhance the accuracy. In the audio modality, mel-frequency cepstral coefficients are extracted for voice-based biometric verification. In the fusion step, features from both modalities are projected into nonlinear Laplacian Eigenmap subspace for multimodal speaker recognition and combined at low level. The proposed approach is tested on the video database of ten human subjects, and the results show that the proposed scheme can attain better accuracy in comparison with the conventional multimodal fusion using latent semantic analysis as well as the single-modality verifications. The experiment on MATLAB shows the potential of the proposed scheme to attain the real-time performance for perceptual HCI applications.

ASR emotional speech: Clarifying the issues and enhancing performance

There are multiple reasons to expect that recognising the verbal content of emotional speech will be a difficult problem, and recognition rates reported in the literature are in fact low. Including information about prosody improves recognition rate for emotions simulated by actors, but its relevance to the freer patterns of spontaneous speech is unproven. This paper shows that recognition rate for spontaneous emotionally coloured speech can be improved by using a language model based on increased representation of emotional utterances. The models are derived by adapting an already existing corpus, the British National Corpus (BNC). An emotional lexicon is used to identify emotionally coloured words, and sentences containing these words are recombined with the BNC to form a corpus with a raised proportion of emotional material. Using a language model based on that technique improves recognition rate by about 20%. (c) 2005 Elsevier Ltd. All rights reserved.

Speaker Recognition In Noisy Conditions With Limited Training Data

In this paper we present a novel method for performing speaker recognition with very limited training data and in the presence of background noise. Similarity-based speaker recognition is considered so that speaker models can be created with limited training speech data. The proposed similarity is a form of cosine similarity used as a distance measure between speech feature vectors. Each speech frame is modelled using subband features, and into this framework, multicondition training and optimal feature selection are introduced, making the system capable of performing speaker recognition in the presence of realistic, time-varying noise, which is unknown during training. Speaker identi?cation experiments were carried out using the SPIDRE database. The performance of the proposed new system for noise compensation is compared to that of an oracle model; the speaker identi?cation accuracy for clean speech by the new system trained with limited training data is compared to that of a GMM trained with several minutes of speech. Both comparisons have demonstrated the effectiveness of the new model. Finally, experiments were carried out to test the new model for speaker identi?cation given limited training data and with differing levels and types of realistic background noise. The results have demonstrated the robustness of the new system.

Illumination invariant facial recognition using a piecewise-constant lighting model

In this paper we demonstrate a simple and novel illumination model that can be used for illumination invariant facial recognition. This model requires no prior knowledge of the illumination conditions and can be used when there is only a single training image per-person. The proposed illumination model separates the effects of illumination over a small area of the face into two components; an additive component modelling the mean illumination and a multiplicative component, modelling the variance within the facial area. Illumination invariant facial recognition is performed in a piecewise manner, by splitting the face image into blocks, then normalizing the illumination within each block based on the new lighting model. The assumptions underlying this novel lighting model have been verified on the YaleB face database. We show that magnitude 2D Fourier features can be used as robust facial descriptors within the new lighting model. Using only a single training image per-person, our new method achieves high (in most cases 100%) identification accuracy on the YaleB, extended YaleB and CMU-PIE face databases.

A Corpus-Based Approach to Speech Enhancement from Nonstationary Noise

Temporal dynamics and speaker characteristics are two important features of speech that distinguish speech from noise. In this paper, we propose a method to maximally extract these two features of speech for speech enhancement. We demonstrate that this can reduce the requirement for prior information about the noise, which can be difficult to estimate for fast-varying noise. Given noisy speech, the new approach estimates clean speech by recognizing long segments of the clean speech as whole units. In the recognition, clean speech sentences, taken from a speech corpus, are used as examples. Matching segments are identified between the noisy sentence and the corpus sentences. The estimate is formed by using the longest matching segments found in the corpus sentences. Longer speech segments as whole units contain more distinct dynamics and richer speaker characteristics, and can be identified more accurately from noise than shorter speech segments. Therefore, estimation based on the longest recognized segments increases the noise immunity and hence the estimation accuracy. The new approach consists of a statistical model to represent up to sentence-long temporal dynamics in the corpus speech, and an algorithm to identify the longest matching segments between the noisy sentence and the corpus sentences. The algorithm is made more robust to noise uncertainty by introducing missing-feature based noise compensation into the corpus sentences. Experiments have been conducted on the TIMIT database for speech enhancement from various types of nonstationary noise including song, music, and crosstalk speech. The new approach has shown improved performance over conventional enhancement algorithms in both objective and subjective evaluations.

Robust bimodal person identification using face and speech with limited training data and corruption of both modalities

This paper presents a novel method of audio-visual fusion for person identification where both the speech and facial modalities may be corrupted, and there is a lack of prior knowledge about the corruption. Furthermore, we assume there is a limited amount of training data for each modality (e.g., a short training speech segment and a single training facial image for each person). A new representation and a modified cosine similarity are introduced for combining and comparing bimodal features with limited training data as well as vastly differing data rates and feature sizes. Optimal feature selection and multicondition training are used to reduce the mismatch between training and testing, thereby making the system robust to unknown bimodal corruption. Experiments have been carried out on a bimodal data set created from the SPIDRE and AR databases with variable noise corruption of speech and occlusion in the face images. The new method has demonstrated improved recognition accuracy.

Perception of Filtered Speech by Children with Developmental Dyslexia and Children with Specific Language Impairments

Here we use two filtered speech tasks to investigate children’s processing of slow (<4 Hz) versus faster (∼33 Hz) temporal modulations in speech. We compare groups of children with either developmental dyslexia (Experiment 1) or speech and language impairments (SLIs, Experiment 2) to groups of typically-developing (TD) children age-matched to each disorder group. Ten nursery rhymes were filtered so that their modulation frequencies were either low-pass filtered (<4 Hz) or band-pass filtered (22 – 40 Hz). Recognition of the filtered nursery rhymes was tested in a picture recognition multiple choice paradigm. Children with dyslexia aged 10 years showed equivalent recognition overall to TD controls for both the low-pass and band-pass filtered stimuli, but showed significantly impaired acoustic learning during the experiment from low-pass filtered targets. Children with oral SLIs aged 9 years showed significantly poorer recognition of band pass filtered targets compared to their TD controls, and showed comparable acoustic learning effects to TD children during the experiment. The SLI samples were also divided into children with and without phonological difficulties. The children with both SLI and phonological difficulties were impaired in recognizing both kinds of filtered speech. These data are suggestive of impaired temporal sampling of the speech signal at different modulation rates by children with different kinds of developmental language disorder. Both SLI and dyslexic samples showed impaired discrimination of amplitude rise times. Implications of these findings for a temporal sampling framework for understanding developmental language disorders are discussed.