Speaker verification based on the fusion of speech acoustics and inverted articulatory signals

Speech perception difficulty modulates theta-band encoding of articulatory synergies

The human brain tracks available speech acoustics and extrapolates missing information such as the speaker's articulatory patterns. However, the extent to which articulatory reconstruction supports speech perception remains unclear. This study explores the relationship between articulatory reconstruction and task difficulty. Participants listened to sentences and performed a speech-rhyming task. Real kinematic data of the speaker's vocal tract were recorded via electromagnetic articulography (EMA) and aligned to corresponding acoustic outputs. We extracted articulatory synergies from the EMA data with principal component analysis (PCA) and employed partial information decomposition (PID) to separate the electroencephalographic (EEG) encoding of acoustic and articulatory features into unique, redundant, and synergistic atoms of information. We median-split sentences into easy (ES) and hard (HS) based on participants' performance and found that greater task difficulty involved greater encoding of unique articulatory information in the theta band. We conclude that fine-grained articulatory reconstruction plays a complementary role in the encoding of speech acoustics, lending further support to the claim that motor processes support speech perception.NEW & NOTEWORTHY Top-down processes originating from the motor system contribute to speech perception through the reconstruction of the speaker's articulatory movement. This study investigates the role of such articulatory simulation under variable task difficulty. We show that more challenging listening tasks lead to increased encoding of articulatory kinematics in the theta band and suggest that, in such situations, fine-grained articulatory reconstruction complements acoustic encoding.

Articulation constrained learning with application to speech emotion recognition

Speech emotion recognition methods combining articulatory information with acoustic features have been previously shown to improve recognition performance. Collection of articulatory data on a large scale may not be feasible in many scenarios, thus restricting the scope and applicability of such methods. In this paper, a discriminative learning method for emotion recognition using both articulatory and acoustic information is proposed. A traditional ℓ ₁-regularized logistic regression cost function is extended to include additional constraints that enforce the model to reconstruct articulatory data. This leads to sparse and interpretable representations jointly optimized for both tasks simultaneously. Furthermore, the model only requires articulatory features during training; only speech features are required for inference on out-of-sample data. Experiments are conducted to evaluate emotion recognition performance over vowels /AA/,/AE/,/IY/,/UW/ and complete utterances. Incorporating articulatory information is shown to significantly improve the performance for valence-based classification. Results obtained for within-corpus and cross-corpus categorical emotion recognition indicate that the proposed method is more effective at distinguishing happiness from other emotions.

The hearing ear is always found close to the speaking tongue: Review of the role of the motor system in speech perception

Does "the motor system" play "a role" in speech perception? If so, where, how, and when? We conducted a systematic review that addresses these questions using both qualitative and quantitative methods. The qualitative review of behavioural, computational modelling, non-human animal, brain damage/disorder, electrical stimulation/recording, and neuroimaging research suggests that distributed brain regions involved in producing speech play specific, dynamic, and contextually determined roles in speech perception. The quantitative review employed region and network based neuroimaging meta-analyses and a novel text mining method to describe relative contributions of nodes in distributed brain networks. Supporting the qualitative review, results show a specific functional correspondence between regions involved in non-linguistic movement of the articulators, covertly and overtly producing speech, and the perception of both nonword and word sounds. This distributed set of cortical and subcortical speech production regions are ubiquitously active and form multiple networks whose topologies dynamically change with listening context. Results are inconsistent with motor and acoustic only models of speech perception and classical and contemporary dual-stream models of the organization of language and the brain. Instead, results are more consistent with complex network models in which multiple speech production related networks and subnetworks dynamically self-organize to constrain interpretation of indeterminant acoustic patterns as listening context requires.