Wireless acoustic sensor networks (WASNs) are formed by a distributed group of acoustic-sensing devices featuring audio playing and recording capabilities. Current mobile computing platforms offer great possibilities for the design of audio-related applications involving acoustic-sensing nodes. In this context, acoustic source localization is one of the application domains that have attracted the most attention of the research community along the last decades. In general terms, the localization of acoustic sources can be achieved by studying energy and temporal and/or directional features from the incoming sound at different microphones and using a suitable model that relates those features with the spatial location of the source (or sources) of interest. This paper reviews common approaches for source localization in WASNs that are focused on different types of acoustic features, namely, the energy of the incoming signals, their time of arrival (TOA) or time difference of arrival (TDOA), the direction of arrival (DOA), and the steered response power (SRP) resulting from combining multiple microphone signals. Additionally, we discuss methods not only aimed at localizing acoustic sources but also designed to locate the nodes themselves in the network. Finally, we discuss current challenges and frontiers in this field.

The imbalance of power supply and demand is an important problem to solve in power industry and Non Intrusive Load Monitoring (NILM) is one of the representative technologies for power demand management. The most critical factor to the NILM is the performance of the classifier among the last steps of the overall NILM operation, and therefore improving the performance of the NILM classifier is an important issue. This paper proposes a new architecture based on the RNN to overcome the limitations of existing classification algorithms and to improve the performance of the NILM classifier. The proposed model, called Multi-Feature Combination Multi-Layer Long Short-Term Memory (MFC-ML-LSTM), adapts various feature extraction techniques that are commonly used for audio signal processing to power signals. It uses Multi-Feature Combination (MFC) for generating the modified input data for improving the classification performance and adopts Multi-Layer LSTM (ML-LSTM) network as the classification model for further improvements. Experimental results show that the proposed method achieves the accuracy and the F1-score for appliance classification with the ranges of 95–100% and 84–100% that are superior to the existing methods based on the Gated Recurrent Unit (GRU) or a single-layer LSTM.

Speaker diarization (SD) is the process of assigning speech segments of an audio stream to its corresponding speakers, thus comprising the problem of voice activity detection (VAD), speaker labeling/identification, and often sound source localization (SSL). Most research activities in the past aimed towards applications as broadcast news, meetings, conversational telephony, and automatic multimodal data annotation, where SD may be performed off-line. However, a recent research focus is human-computer interaction (HCI) systems where SD must be performed on-line, and in real-time, as in modern gaming devices and interaction with large displays. Often, such applications further suffer from noise, reverberations, and overlapping speech, making them increasingly challenging. In such situations, multimodal/multisensory approaches can provide more accurate results than unimodal ones, given a data stream may compensate for occasional instabilities of other modalities. Accordingly, this paper presents an on-line multimodal SD algorithm designed to work in a realistic environment with multiple, overlapping speakers. Our work employs a microphone array, a color camera, and a depth sensor as input streams, from which speech-related features are extracted to be later merged through a support vector machine approach consisting of VAD and SSL modules. Speaker identification is incorporated through a hybrid technique of face positioning history and face recognition. Our final SD approach experimentally achieves an average diarization error rate of 11.48% in scenarios with up to three simultaneous speakers, and is able to run 3.2 × real-time.

Automatic peer-to-peer energy trading can be defined as a Markov decision process and designed using deep reinforcement learning. We consider prosumer as an entity that consumes and produces electric energy with an energy storage system, and define the prosumer’s objective as maximizing the profit through participation in peer-to-peer energy trading, similar to that of the agents in stock trading. In this paper, we propose an automatic peer-to-peer energy trading model by adopting a deep Q-network-based automatic trading algorithm originally designed for stock trading. Unlike in stock trading, the assets held by a prosumer may change owing to factors such as the consumption and generation of energy by the prosumer in addition to the changes from trading activities. Therefore, we propose a new trading evaluation criterion that considers these factors by defining profit as the sum of the gains from four components: electricity bill, trading, electric energy stored in the energy storage system, and virtual loss. For the proposed automatic peer-to-peer energy trading algorithm, we adopt a long-term delayed reward method that evaluates the delayed reward that occurs once per month by generating the termination point of an episode at each month and propose a long short-term delayed reward method that compensates for the issue with the long-term delayed reward method having only a single evaluation per month. This long short-term delayed reward method enables effective learning of the monthly long-term trading patterns and the short-term trading patterns at the same time, leading to a better trading strategy. The experimental results showed that the long short-term delayed reward method-based energy trading model achieves higher profits every month both in the progressive and fixed rate systems throughout the year and that prosumer participating in the trading not only earns profits every month but also reduces loss from over-generation of electric energy in the case of South Korea. Further experiments with various progressive rate systems of Japan, Taiwan, and the United States as well as in different prosumer environments indicate the general applicability of the proposed method.

Along with automatic speech recognition, many researchers have been actively studying speech emotion recognition, since emotion information is as crucial as the textual information for effective interactions. Emotion can be divided into categorical emotion and dimensional emotion. Although categorical emotion is widely used, dimensional emotion, typically represented as arousal and valence, can provide more detailed information on the emotional states. Therefore, in this paper, we propose a Conformer-based model for arousal and valence recognition. Our model uses Conformer as an encoder, a fully connected layer as a decoder, and statistical pooling layers as a connector. In addition, we adopted multi-task learning and multi-feature combination, which showed a remarkable performance for speech emotion recognition and time-series analysis, respectively. The proposed model achieves a state-of-the-art recognition accuracy of 70.0 ± 1.5% for arousal in terms of unweighted accuracy on the IEMOCAP dataset.