The use of speech and lip modalities for robust speaker verification under adverse conditions

Investigates the use of lip information, in conjunction with speech information, for robust speaker verification in the presence of background noise. We have previously shown (Int. Conf. on Acoustics, Speech and Signal Proc., vol. 6, pp. 3693-3696, May 1998) that features extracted from a speaker's moving lips hold speaker dependencies which are complementary with speech features. We demonstrate that the fusion of lip and speech information allows for a highly robust speaker verification system which outperforms either subsystem individually. We present a new technique for determining the weighting to be applied to each modality so as to optimize the performance of the fused system. Given a correct weighting, lip information is shown to be highly effective for reducing the false acceptance and false rejection error rates in the presence of background noise

Robust speaker verification via fusion of speech and lip modalities

This paper investigates the use of lip information, in conjunction with speech information, for robust speaker verification in the presence of background noise. It has been previously shown in our own work, and in the work of others, that features extracted from a speaker's moving lips hold speaker dependencies which are complementary with speech features. We demonstrate that the fusion of lip and speech information allows for a highly robust speaker verification system which outperforms the performance of either sub-system. We present a new technique for determining the weighting to be applied to each modality so as to optimize the performance of the fused system. Given a correct weighting, lip information is shown to be highly effective for reducing the false acceptance and false rejection error rates in the presence of background noise

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.

Improving visual noise insensitivity in small vocabulary audio visual speech recognition applications

Visual noise insensitivity is important to audio visual speech recognition (AVSR). Visual noise can take on a number of forms such as varying frame rate, occlusion, lighting or speaker variabilities. The use of a high dimensional secondary classifier on the word likelihood scores from both the audio and video modalities is investigated for the purposes of adaptive fusion. Preliminary results are presented demonstrating performance above the catastrophic fusion boundary for our confidence measure irrespective of the type of visual noise presented to it. Our experiments were restricted to small vocabulary applications.

Real-time magnetic resonance imaging and electromagnetic articulography database for speech production research (TC)

USC-TIMIT is an extensive database of multimodal speech production data, developed to complement existing resources available to the speech research community and with the intention of being continuously refined and augmented. The database currently includes real-time magnetic resonance imaging data from five male and five female speakers of American English. Electromagnetic articulography data have also been presently collected from four of these speakers. The two modalities were recorded in two independent sessions while the subjects produced the same 460 sentence corpus used previously in the MOCHA-TIMIT database. In both cases the audio signal was recorded and synchronized with the articulatory data. The database and companion software are freely available to the research community. (C) 2014 Acoustical Society of America.

Robust Audio-Visual Speech Recognition under Noisy Audio-Video Conditions

This paper presents the maximum weighted stream posterior (MWSP) model as a robust and efficient stream integration method for audio-visual speech recognition in environments, where the audio or video streams may be subjected to unknown and time-varying corruption. A significant advantage of MWSP is that it does not require any specific measurements of the signal in either stream to calculate appropriate stream weights during recognition, and as such it is modality-independent. This also means that MWSP complements and can be used alongside many of the other approaches that have been proposed in the literature for this problem. For evaluation we used the large XM2VTS database for speaker-independent audio-visual speech recognition. The extensive tests include both clean and corrupted utterances with corruption added in either/both the video and audio streams using a variety of types (e.g., MPEG-4 video compression) and levels of noise. The experiments show that this approach gives excellent performance in comparison to another well-known dynamic stream weighting approach and also compared to any fixed-weighted integration approach in both clean conditions or when noise is added to either stream. Furthermore, our experiments show that the MWSP approach dynamically selects suitable integration weights on a frame-by-frame basis according to the level of noise in the streams and also according to the naturally fluctuating relative reliability of the modalities even in clean conditions. The MWSP approach is shown to maintain robust recognition performance in all tested conditions, while requiring no prior knowledge about the type or level of noise.

A new comparator account of auditory verbal hallucinations: how motor prediction can plausibly contribute to the sense of agency for inner speech

The comparator account holds that processes of motor prediction contribute to the sense of agency by attenuating incoming sensory information and that disruptions to this process contribute to misattributions of agency in schizophrenia. Over the last 25 years this simple and powerful model has gained widespread support not only as it relates to bodily actions but also as an account of misattributions of agency for inner speech, potentially explaining the etiology of auditory verbal hallucination (AVH). In this paper we provide a detailed analysis of the traditional comparator account for inner speech, pointing out serious problems with the specification of inner speech on which it is based and highlighting inconsistencies in the interpretation of the electrophysiological evidence commonly cited in its favor. In light of these analyses we propose a new comparator account of misattributed inner speech. The new account follows leading models of motor imagery in proposing that inner speech is not attenuated by motor prediction, but rather derived directly from it. We describe how failures of motor prediction would therefore directly affect the phenomenology of inner speech and trigger a mismatch in the comparison between motor prediction and motor intention, contributing to abnormal feelings of agency. We argue that the new account fits with the emerging phenomenological evidence that AVHs are both distinct from ordinary inner speech and heterogeneous. Finally, we explore the possibility that the new comparator account may extend to explain disruptions across a range of imagistic modalities, and outline avenues for future research.

End-tidal CO(2). An important parameter for a correct interpretation in functional brain studies using speech tasks

The aim was to investigate the effect of different speech tasks, i.e. recitation of prose (PR), alliteration (AR) and hexameter (HR) verses and a control task (mental arithmetic (MA) with voicing of the result on end-tidal CO2 (PETCO2), cerebral hemodynamics and oxygenation. CO2 levels in the blood are known to strongly affect cerebral blood flow. Speech changes breathing pattern and may affect CO2 levels. Measurements were performed on 24 healthy adult volunteers during the performance of the 4 tasks. Tissue oxygen saturation (StO2) and absolute concentrations of oxyhemoglobin ([O2Hb]), deoxyhemoglobin ([HHb]) and total hemoglobin ([tHb]) were measured by functional near-infrared spectroscopy (fNIRS) and PETCO2 by a gas analyzer. Statistical analysis was applied to the difference between baseline before the task, 2 recitation and 5 baseline periods after the task. The 2 brain hemispheres and 4 tasks were tested separately. A significant decrease in PETCO2 was found during all 4 tasks with the smallest decrease during the MA task. During the recitation tasks (PR, AR and HR) a statistically significant (p < 0.05) decrease occurred for StO2 during PR and AR in the right prefrontal cortex (PFC) and during AR and HR in the left PFC. [O2Hb] decreased significantly during PR, AR and HR in both hemispheres. [HHb] increased significantly during the AR task in the right PFC. [tHb] decreased significantly during HR in the right PFC and during PR, AR and HR in the left PFC. During the MA task, StO2 increased and [HHb] decreased significantly during the MA task. We conclude that changes in breathing (hyperventilation) during the tasks led to lower CO2 pressure in the blood (hypocapnia), predominantly responsible for the measured changes in cerebral hemodynamics and oxygenation. In conclusion, our findings demonstrate that PETCO2 should be monitored during functional brain studies investigating speech using neuroimaging modalities, such as fNIRS, fMRI to ensure a correct interpretation of changes in hemodynamics and oxygenation.

The Neural Substrates of Multisensory Speech Perception

Comprehending speech is one of the most important human behaviors, but we are only beginning to understand how the brain accomplishes this difficult task. One key to speech perception seems to be that the brain integrates the independent sources of information available in the auditory and visual modalities in a process known as multisensory integration. This allows speech perception to be accurate, even in environments in which one modality or the other is ambiguous in the context of noise. Previous electrophysiological and functional magnetic resonance imaging (fMRI) experiments have implicated the posterior superior temporal sulcus (STS) in auditory-visual integration of both speech and non-speech stimuli. While evidence from prior imaging studies have found increases in STS activity for audiovisual speech compared with unisensory auditory or visual speech, these studies do not provide a clear mechanism as to how the STS communicates with early sensory areas to integrate the two streams of information into a coherent audiovisual percept. Furthermore, it is currently unknown if the activity within the STS is directly correlated with strength of audiovisual perception. In order to better understand the cortical mechanisms that underlie audiovisual speech perception, we first studied the STS activity and connectivity during the perception of speech with auditory and visual components of varying intelligibility. By studying fMRI activity during these noisy audiovisual speech stimuli, we found that STS connectivity with auditory and visual cortical areas mirrored perception; when the information from one modality is unreliable and noisy, the STS interacts less with the cortex processing that modality and more with the cortex processing the reliable information. We next characterized the role of STS activity during a striking audiovisual speech illusion, the McGurk effect, to determine if activity within the STS predicts how strongly a person integrates auditory and visual speech information. Subjects with greater susceptibility to the McGurk effect exhibited stronger fMRI activation of the STS during perception of McGurk syllables, implying a direct correlation between strength of audiovisual integration of speech and activity within an the multisensory STS.

Effects of inner and heard speech in arts speech therapy on cerebral oxygenation and hemodynamics

Background: The aim of the present study was to contributing to researching physiological effects of arts speech therapy by (i) investigating effects of inner and heard speech on cerebral hemodynamics and oxygenation, and (ii) analyzing if these changes were affected by alterations of the arterial carbon dioxide pressure (PaCO2). Methods: In 29 healthy adult volunteers we measured changes in cerebral absolute oxyhemoglobin ([O2Hb]), deoxyhemoglobin ([HHb]), total hemoglobin ([tHb]) concentrations and tissue oxygen saturation (StO2) (over the left and right anterior prefrontal cortex (PFC)) using functional near-infrared spectroscopy (fNIRS) as well as changes in end-tidal CO2 (PETCO2) using capnography. Each subject performed six different tasks: three types of task modalities, i.e. inner speech, heard speech from a person and heard speech from a record, and, two recitation texts, i.e. hexameter and alliteration on different days according to a randomized crossover design. Statistical analysis was applied to the differences between the baseline, two task and four recovery periods. The two brain hemispheres, i.e. left and right PFC, and six tasks were tested separately. Results: During the tasks we found in general a decrease in PETCO2 (significantly only for inner speech), StO2, [O2Hb], [tHb] as well as in an increase in [HHb]. There was a significant difference between hexameter and alliteration. Particularly, the changes in [tHb] at the left PFC during tasks and after them were statistically different. Furthermore we found significant relations between changes in [O2Hb], [HHb], [tHb] or StO2 and the participants’ age, the baseline PETCO2, or certain speech tasks. Conclusions: Changes in breathing (hyperventilation) during the tasks led to lower PaCO2 (hypocapnia) for inner speech. During heard speech no significant changes in PaCO2 occurred, but the decreases in StO2, [O2Hb], [tHb] suggest that changes in PaCO2 were also relevant here. Different verse types (hexameter, alliteration) led to different changes in [tHb]. Consequently, StO2, [O2Hb], [HHb] and [tHb] are affected by interplay of both PaCO2 reactivity and task dependent functional brain activity.

Improving Speech Interaction in Vehicles Using Context-Aware Information through A SCXML Framework

Speech Technologies can provide important benefits for the development of more usable and safe in-vehicle human-machine interactive systems (HMIs). However mainly due robustness issues, the use of spoken interaction can entail important distractions to the driver. In this challenging scenario, while speech technologies are evolving, further research is necessary to explore how they can be complemented with both other modalities (multimodality) and information from the increasing number of available sensors (context-awareness). The perceived quality of speech technologies can significantly be increased by implementing such policies, which simply try to make the best use of all the available resources; and the in vehicle scenario is an excellent test-bed for this kind of initiatives. In this contribution we propose an event-based HMI design framework which combines context modelling and multimodal interaction using a W3C XML language known as SCXML. SCXML provides a general process control mechanism that is being considered by W3C to improve both voice interaction (VoiceXML) and multimodal interaction (MMI). In our approach we try to anticipate and extend these initiatives presenting a flexible SCXML-based approach for the design of a wide range of multimodal context-aware HMI in-vehicle interfaces. The proposed framework for HMI design and specification has been implemented in an automotive OSGi service platform, and it is being used and tested in the Spanish research project MARTA for the development of several in-vehicle interactive applications.

Infants' selective attention to faces and prosody of speech: The roles of intersensory redundancy and exploratory time

One of the overarching questions in the field of infant perceptual and cognitive development concerns how selective attention is organized during early development to facilitate learning. The following study examined how infants' selective attention to properties of social events (i.e., prosody of speech and facial identity) changes in real time as a function of intersensory redundancy (redundant audiovisual, nonredundant unimodal visual) and exploratory time. Intersensory redundancy refers to the spatially coordinated and temporally synchronous occurrence of information across multiple senses. Real time macro- and micro-structural change in infants' scanning patterns of dynamic faces was also examined. ^ According to the Intersensory Redundancy Hypothesis, information presented redundantly and in temporal synchrony across two or more senses recruits infants' selective attention and facilitates perceptual learning of highly salient amodal properties (properties that can be perceived across several sensory modalities such as the prosody of speech) at the expense of less salient modality specific properties. Conversely, information presented to only one sense facilitates infants' learning of modality specific properties (properties that are specific to a particular sensory modality such as facial features) at the expense of amodal properties (Bahrick & Lickliter, 2000, 2002). ^ Infants' selective attention and discrimination of prosody of speech and facial configuration was assessed in a modified visual paired comparison paradigm. In redundant audiovisual stimulation, it was predicted infants would show discrimination of prosody of speech in the early phases of exploration and facial configuration in the later phases of exploration. Conversely, in nonredundant unimodal visual stimulation, it was predicted infants would show discrimination of facial identity in the early phases of exploration and prosody of speech in the later phases of exploration. Results provided support for the first prediction and indicated that following redundant audiovisual exposure, infants showed discrimination of prosody of speech earlier in processing time than discrimination of facial identity. Data from the nonredundant unimodal visual condition provided partial support for the second prediction and indicated that infants showed discrimination of facial identity, but not prosody of speech. The dissertation study contributes to the understanding of the nature of infants' selective attention and processing of social events across exploratory time.^