Audio classification is an important task of mapping audio samples into their corresponding labels. Recently, the transformer model with self-attention mechanisms has been adopted in this field. However, existing audio transformers require large GPU memories and long training time, meanwhile relying on pretrained vision models to achieve high performance, which limits the model’s scalability in audio tasks. To combat these problems, we introduce HTS-AT: an audio transformer with a hierarchical structure to reduce the model size and training time. It is further combined with a token-semantic module to map final outputs into class featuremaps, thus enabling the model for the audio event detection (i.e. localization in time). We evaluate HTS-AT on three datasets of audio classification where it achieves new state-of-the-art (SOTA) results on AudioSet and ESC50, and equals the SOTA on Speech Command V2. It also achieves better performance in event localization than the previous CNN-based models. Moreover, HTS-AT requires only 35% model parameters and 15% training time of the previous audio transformer. These results demonstrate the high performance and high efficiency of HTS-AT.

We present an information-theoretic measure for the amount of randomness or stochasticity that exists in a signal. This measure is formulated in terms of the rate of growth of multi-information for every new signal sample of the signal that is observed over time. In case of a Gaussian statistics it is shown that this measure is equivalent to the well-known spectral flatness measure that is commonly used in audio processing. For nonGaussian linear processes a generalized spectral flatness measure is developed, which estimates the excessive structure that is present in the signal due to the nonGaussianity of the innovation process. An estimator for this measure is developed using Negentropy approximation to the non-Gaussian signal and the innovation process statistics. Applications of this new measure are demonstrated for the problem of voiced/unvoiced determination, showing improved performance.

We present an exhaustive review of research on automatic classification of sounds from musical instruments. Two different but complementary approaches are examined, the perceptual approach and the taxonomic approach. The former is targeted to derive perceptual similarity functions in order to use them for timbre clustering and for searching and retrieving sounds by timbral similarity. The latter is targeted to derive indexes for labeling sounds after culture- or user-biased taxonomies. We review the relevant features that have been used in the two areas and then we present and discuss different techniques for similarity-based clustering of sounds and for classification into pre-defined instrumental categories.

Modern speech processing applications require operation on signal of interest that is contaminated by high level of noise. This situation calls for a greater robustness in estimation of the speech parameters, a task which is hard to achieve using standard speech models. In this paper, we present an optimal estimation procedure for sound signals (such as speech) that are modeled by harmonic sources. The harmonic model achieves more robust and accurate estimation of voiced speech parameters. Using maximum a posteriori probability framework, successful tracking of pitch parameters is possible in ultra low signal to noise conditions (as low as -15 dB). The performance of the method is evaluated using the Keele pitch detection database with realistic background noise. The results show best performance in comparison to other state-of-the-art pitch detectors. Application of the proposed algorithm in a simple speaker identification system shows significant improvement in the performance.