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.

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.

Robust indoor speaker recognition in a network of audio and video sensors

Situational awareness is achieved naturally by the human senses of sight and hearing in combination. Automatic scene understanding aims at replicating this human ability using microphones and cameras in cooperation. In this paper, audio and video signals are fused and integrated at different levels of semantic abstractions. We detect and track a speaker who is relatively unconstrained, i.e., free to move indoors within an area larger than the comparable reported work, which is usually limited to round table meetings. The system is relatively simple: consisting of just 4 microphone pairs and a single camera. Results show that the overall multimodal tracker is more reliable than single modality systems, tolerating large occlusions and cross-talk. System evaluation is performed on both single and multi-modality tracking. The performance improvement given by the audio–video integration and fusion is quantified in terms of tracking precision and accuracy as well as speaker diarisation error rate and precision–recall (recognition). Improvements vs. the closest works are evaluated: 56% sound source localisation computational cost over an audio only system, 8% speaker diarisation error rate over an audio only speaker recognition unit and 36% on the precision–recall metric over an audio–video dominant speaker recognition method.

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.

Feature Selection for Pose Invariant Lip Biometrics

For the first time in this paper the authors present results showing the effect of out of plane speaker head pose variation on a lip biometric based speaker verification system. Using appearance DCT based features, they adopt a Mutual Information analysis technique to highlight the class discriminant DCT components most robust to changes in out of plane pose. Experiments are conducted using the initial phase of a new multi view Audio-Visual database designed for research and development of pose-invariant speech and speaker recognition. They show that verification performance can be improved by substituting higher order horizontal DCT components for vertical, particularly in the case of a train/test pose angle mismatch.

Robust Multimodal Person Identification With Limited Training Data

This paper presents a novel method of audio-visual feature-level 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 are 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 multimodal feature representation and a modified cosine similarity are introduced to combine and compare 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 dataset created from the SPIDRE speaker recognition database and AR face recognition database with variable noise corruption of speech and occlusion in the face images. The system's speaker identification performance on the SPIDRE database, and facial identification performance on the AR database, is comparable with the literature. Combining both modalities using the new method of multimodal fusion leads to significantly improved accuracy over the unimodal systems, even when both modalities have been corrupted. The new method also shows improved identification accuracy compared with the bimodal systems based on multicondition model training or missing-feature decoding alone.

Comparison of Image Transform-Based Features for Visual Speech Recognition in Clean and Corrupted Videos

We present results of a study into the performance of a variety of different image transform-based feature types for speaker-independent visual speech recognition of isolated digits. This includes the first reported use of features extracted using a discrete curvelet transform. The study will show a comparison of some methods for selecting features of each feature type and show the relative benefits of both static and dynamic visual features. The performance of the features will be tested on both clean video data and also video data corrupted in a variety of ways to assess each feature type's robustness to potential real-world conditions. One of the test conditions involves a novel form of video corruption we call jitter which simulates camera and/or head movement during recording.

Hidden Conditional Random Fields for Visual Speech Recognition

In this paper we present the application of Hidden Conditional Random Fields (HCRFs) to modelling speech for visual speech recognition. HCRFs may be easily adapted to model long range dependencies across an observation sequence. As a result visual word recognition performance can be improved as the model is able to take more of a contextual approach to generating state sequences. Results are presented from a speaker-dependent, isolated digit, visual speech recognition task using comparisons with a baseline HMM system. We firstly illustrate that word recognition rates on clean video using HCRFs can be improved by increasing the number of past and future observations being taken into account by each state. Secondly we compare model performances using various levels of video compression on the test set. As far as we are aware this is the first attempted use of HCRFs for visual speech recognition.

Inter-Frame Contextual Modelling For Visual Speech Recognition

In this paper, we present a new approach to visual speech recognition which improves contextual modelling by combining Inter-Frame Dependent and Hidden Markov Models. This approach captures contextual information in visual speech that may be lost using a Hidden Markov Model alone. We apply contextual modelling to a large speaker independent isolated digit recognition task, and compare our approach to two commonly adopted feature based techniques for incorporating speech dynamics. Results are presented from baseline feature based systems and the combined modelling technique. We illustrate that both of these techniques achieve similar levels of performance when used independently. However significant improvements in performance can be achieved through a combination of the two. In particular we report an improvement in excess of 17% relative Word Error Rate in comparison to our best baseline system.

FPGA Implementation of a Pipelined Gaussian Calculation for HMM-Based Large Vocabulary Speech Recognition

A scalable large vocabulary, speaker independent speech recognition system is being developed using Hidden Markov Models (HMMs) for acoustic modeling and a Weighted Finite State Transducer (WFST) to compile sentence, word, and phoneme models. The system comprises a software backend search and an FPGA-based Gaussian calculation which are covered here. In this paper, we present an efficient pipelined design implemented both as an embedded peripheral and as a scalable, parallel hardware accelerator. Both architectures have been implemented on an Alpha Data XRC-5T1, reconfigurable computer housing a Virtex 5 SX95T FPGA. The core has been tested and is capable of calculating a full set of Gaussian results from 3825 acoustic models in 9.03 ms which coupled with a backend search of 5000 words has provided an accuracy of over 80%. Parallel implementations have been designed with up to 32 cores and have been successfully implemented with a clock frequency of 133?MHz.