Search

Owing to the suprasegmental behavior of emotional speech, turn-level features have demonstrated a better success than frame-level features for recognition-related tasks. Conventionally, such features are obtained via a brute-force collection of statistics over frames, thereby losing important local information in the process which affects the performance. To overcome these limitations, a novel feature extraction approach using latent topic models (LTMs) is presented in this study. Speech is assumed to comprise of a mixture of emotion-specific topics, where the latter capture emotionally salient information from the co-occurrences of frame-level acoustic features and yield better descriptors. Specifically, a supervised replicated softmax model (sRSM), based on restricted Boltzmann machines and distributed representations, is proposed to learn naturally discriminative topics. The proposed features are evaluated for the recognition of categorical or continuous emotional attributes via within and cross-corpus experiments conducted over acted and spontaneous expressions. In a within-corpus scenario, sRSM outperforms competing LTMs, while obtaining a significant improvement of 16.75% over popular statistics-based turn-level features for valence-based classification, which is considered to be a difficult task using only speech. Further analyses with respect to the turn duration show that the improvement is even more significant, 35%, on longer turns (>6 s), which is highly desirable for current turn-based practices. In a cross-corpus scenario, two novel adaptation-based approaches, instance selection, and weight regularization are proposed to reduce the inherent bias due to varying annotation procedures and cultural perceptions across databases. Experimental results indicate a natural, yet less severe, deterioration in performance - only 2.6% and 2.7%, thereby highlighting the generalization ability of the proposed features. [ABSTRACT FROM AUTHOR]

Environmental monitoring relies on compact, portable sensor systems capable of detecting pollutants in real-time. An integrated chemical sensor array system is developed for detection and identification of environmental pollutants in diesel and gasoline exhaust fumes. The system consists of a low noise floor analog front-end (AFE) followed by a signal processing stage. In this paper, we present techniques to detect, digitize, denoise and classify a certain set of analytes. The proposed AFE reads out the output of eight conductometric sensors and eight amperometric electrochemical sensors and achieves 91 dB SNR at 23.4 mW quiescent power consumption for all channels. We demonstrate signal denoising using a discrete wavelet transform based technique. Appropriate features are extracted from sensor data, and pattern classification methods are used to identify the analytes. Several existing pattern classification algorithms are used for analyte detection and the comparative results are presented. [ABSTRACT FROM AUTHOR]

Organizing a database of user-contributed environmental sound recordings allows sound files to be linked not only by the semantic tags and labels applied to them, but also to other sounds with similar acoustic characteristics. Of paramount importance in navigating these databases are the problems of retrieving similar sounds using text- or sound-based queries, and automatically annotating unlabeled sounds. We propose an integrated system, which can be used for text-based retrieval of unlabeled audio, content-based query-by-example, and automatic annotation of unlabeled sound files. To this end, we introduce an ontological framework where sounds are connected to each other based on the similarity between acoustic features specifically adapted to environmental sounds, while semantic tags and sounds are connected through link weights that are optimized based on user provided tags. Furthermore, tags are linked to each other through a measure of semantic similarity, which allows for efficient incorporation of out-of-vocabulary tags, that is, tags that do not yet exist in the database. Results on two freely available databases of environmental sounds contributed and labeled by non expert users demonstrate effective recall, precision, and average precision scores for both the text-based retrieval and annotation tasks. [ABSTRACT FROM AUTHOR]

Many modern speech bandwidth extension techniques predict the high-frequency band based on features extracted fromthe lower band.While this method works for certain types of speech, problems arise when the correlation between the low and the high bands is not sufficient for adequate prediction. These situations require that additional high-band information is sent to the decoder. This overhead information, however, can be cleverly quantized using human auditory system models. In this paper, we propose a novel speech compression method that relies on bandwidth extension. The novelty of the technique lies in an elaborate perceptualmodel that determines a quantization scheme for wideband recovery and synthesis. Furthermore, a source/filter bandwidth extension algorithm based on spectral spline fitting is proposed. Results reveal that the proposed system improves the quality of narrowband speech while performing at a lower bitrate. When compared to other wideband speech coding schemes, the proposed algorithms provide comparable speech quality at a lower bitrate. [ABSTRACT FROM AUTHOR]

Effective patient care mandates rapid, yet accurate, diagnosis. With the abundance of non-invasive diagnostic measurements and electronic health records (EHR), manual interpretation for differential diagnosis has become time-consuming and challenging. This has led to wide-spread adoption of AI-powered tools, in pursuit of improving accuracy and efficiency of this process. While the unique challenges presented by each modality and clinical task demand customized tools, the cumbersome process of making problem-specific choices has triggered the critical need for a generic solution to enable rapid development of models in practice. In this spirit, we develop DDxNet, a deep architecture for time-varying clinical data, which we demonstrate to be well-suited for diagnostic tasks involving different modalities (ECG/EEG/EHR), required level of characterization (abnormality detection/phenotyping) and data fidelity (single-lead ECG/22-channel EEG). Using multiple benchmark problems, we show that DDxNet produces high-fidelity predictive models, and sometimes even provides significant performance gains over problem-specific solutions. [ABSTRACT FROM AUTHOR]

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 ℓ1-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. [ABSTRACT FROM AUTHOR]