Environmental sound classification is an important branch of acoustic signal processing. In this work, a set of sound classification features based on audio signal perception and statistical analysis are proposed to describe the signal from multiple aspects of the time and frequency domain. Energy features, spectral entropy features, zero crossing rate (ZCR), and mel-frequency cepstral coefficient (MFCC) are combined to form joint signal analysis (JSA) features to improve the signal expression of the features. Then, based on the JSA, a novel region joint signal analysis feature (RJSA) for environment sound classification is also proposed. It can reduce feature extraction computation and improve feature stability, robustness, and classification accuracy. Finally, a sound classification framework based on the boosting ensemble learning method is provided to improve the classification accuracy and model generalization. The experimental results show that compared with the highest classification accuracy of the baseline algorithm, the environmental sound classification algorithm based on our proposed RJSA features and ensemble learning methods improves the classification accuracy, and the accuracy of the LightGBM-based sound classification algorithm improves by 14.6%.

As the memory footprint requirement and computational scale concerned, the light-weighted Binary Neural Networks (BNNs) have great advantages in limited-resources platforms, such as AIoT (Artificial Intelligence in Internet of Things) edge terminals, wearable and portable devices, etc. However, the binarization process naturally brings considerable information losses and further deteriorates the accuracy. In this article, three aspects are introduced to better the binarized ReActNet accuracy performance with a more low-complex computation. Firstly, an improved Binarized Ghost Module (BGM) for the ReActNet is proposed to increase the feature maps information. At the same time, the computational scale of this structure is still kept at a very low level. Secondly, we propose a new Label-aware Loss Function (LLF) in the penultimate layer as a supervisor which takes the label information into consideration. This auxiliary loss function makes each category’s feature vectors more separate, and improve the final fully-connected layer’s classification accuracy accordingly. Thirdly, the Normalization-based Attention Module (NAM) method is adopted to regulate the activation flow. The module helps to avoid the gradient saturation problem. With these three approaches, our improved binarized network outperforms the other state-of-the-art methods. It can achieve 71.4% Top-1 accuracy on the ImageNet and 86.45% accuracy on the CIFAR-10 respectively. Meanwhile, its computational scale OPs is the least 0.86×10⁸ compared with the other mainstream BNN models. The experimental results prove the effectiveness of our proposals, and the study is very helpful and promising for the future low-power hardware implementations.

In terms of memory footprint requirement and computing speed, the binary neural networks (BNNs) have great advantages in power-aware deployment applications, such as AIoT edge terminals, wearable and portable devices, etc. However, the networks’ binarization process inevitably brings considerable information losses, and further leads to accuracy deterioration. To tackle these problems, we initiate analyzing from a perspective of the information theory, and manage to improve the networks information capacity. Based on the analyses, our work has two primary contributions: the first is a newly proposed median loss (ML) regularization technique. It improves the binary weights distribution more evenly, and consequently increases the information capacity of BNNs greatly. The second is the batch median of activations (BMA) method. It raises the entropy of activations by subtracting a median value, and simultaneously lowers the quantization error by computing separate scaling factors for the positive and negative activations procedure. Experiment results prove that the proposed methods utilized in ResNet-18 and ResNet-34 individually outperform the Bi-Real baseline by 1.3% and 0.9% Top-1 accuracy on the ImageNet 2012. Proposed ML and BMA for the storage cost and calculation complexity increments are minor and negligible. Additionally, comprehensive experiments also prove that our methods can be applicable and embedded into the present popular BNN networks with accuracy improvement and negligible overhead increment.

Linear prediction is the kernel technology in speech processing. It has been widely applied in speech recognition, synthesis, and coding, and can efficiently and correctly represent the speech frequency spectrum with only a few parameters. Line Spectrum Pairs (LSPs) frequencies, as an alternative representation of Linear Predictive Coding (LPC), have the advantages of good quantization accuracy and low spectral sensitivity. However, computing the LSPs frequencies takes a long time. To address this issue, a fast computation algorithm, based on the Bairstow method for computing LSPs frequencies from linear prediction coefficients, is proposed in this paper. The algorithm process first transforms the symmetric and antisymmetric polynomial to general polynomial, then extracts the polynomial roots. Associated with the short-term stationary property of speech signal, an adaptive initial method is applied to reduce the average iteration numbers by 26%, as compared to the statics in the initial method, with a Perceptual Evaluation of Speech Quality (PESQ) score reaching 3.46. Experimental results show that the proposed method can extract the polynomial roots efficiently and accurately with significantly reduced computation complexity. Compared to previous works, the proposed method is 17 times faster than Tschirnhus Transform, and has a 22% PESQ improvement on the Birge-Vieta method with an almost comparable computation time.

A high-performance vector quantization (VQ) codebook search algorithm is proposed in this paper. VQ is an important data compression technique that has been widely applied to speech, image, and video compression. However, the process of the codebook search demands a high computational load. To solve this issue, a novel algorithm that consists of training and encoding procedures is proposed. In the training procedure, a training speech dataset was used to build the squared-error distortion look-up table for each subspace. In the encoding procedure, firstly, an input vector was quickly assigned to a search subspace. Secondly, the candidate code word group was obtained by employing the triangular inequality elimination (TIE) equation. Finally, a partial distortion elimination technique was employed to reduce the number of multiplications. The proposed method reduced the number of searches and computation load significantly, especially when the input vectors were uncorrelated. The experimental results show that the proposed algorithm provides a computational saving (CS) of up to 85% in the full search algorithm, up to 76% in the TIE algorithm, and up to 63% in the iterative TIE algorithm. Further, the proposed method provides CS and load reduction of up to 29–33% and 67–69%, respectively, over the BSS-ITIE algorithm.

Achieving coherent sampling has always been a major challenge in analog-to-digital converter (ADC) spectral testing. If the coherent sampling condition cannot be met, leakage appears in the spectrum, which results in inaccurate parameters. The windowing method is widely used to process ADC data to eliminate leakage. However, this method requires prior knowledge about the type of window, and some commonly used windows cannot provide accurate results for a high precision ADC. In this paper, some general principles for optimizing the cosine-sum window and some windows with maximum side-lobe decay have been presented. A test method for eliminating the leakage caused by non-coherent sampling is also proposed. The proposed method can accurately evaluate the dynamic parameters of an ADC with arbitrary non-coherency. Various simulation results and measurement data demonstrate the functionality and robustness of the proposed method. The proposed method significantly relaxes the condition of coherent sampling and decreases the test cost.