The presence of laryngeal disease affects vocal fold(s) dynamics and thus causes changes in pitch, loudness, and other characteristics of the human voice. Many frameworks based on the acoustic analysis of speech signals have been created in recent years; however, they are evaluated on just one or two corpora and are not independent to voice illnesses and human bias. In this article, a unified wavelet-based paradigm for evaluating voice diseases is presented. This approach is independent of voice diseases, human bias, or dialect. The vocal folds’ dynamics are impacted by the voice disorder, and this further modifies the sound source. Therefore, inverse filtering is used to capture the modified voice source. Furthermore, the fundamental frequency independent statistical and energy metrics are derived from each spectral sub-band to characterize the retrieved voice source. Speech recordings of the sustained vowel /a/ were collected from four different datasets in German, Spanish, English, and Arabic to run the several intra and inter-dataset experiments. The classifiers’ achieved performance indicators show that energy and statistical features uncover vital information on a variety of clinical voices, and therefore the suggested approach can be used as a complementary means for the automatic medical assessment of voice diseases.
This article proposes an ambulance vehicle routing approach in smart cities. The approach is based on the bat algorithm and convolutional neural network (BA-CNN). It aims to take transfer the patients confidentially, accurately, and quickly. The type of CNN used in this research is a residual network (ResNet). The node method is responsible for creating the city map. In the beginning, information about the accident place is received by the control station and forwarded to both the hospital and the ambulance. The driver feeds the data that contain the ambulance vehicle’s node position and the accident location to the BA-CNN vehicle routing algorithm. The algorithm then obtains the shortest path to reach the location of the accident by the driver. When the vehicle arrives at the accident location, the driver updates the algorithm with hospital and accident positions. Then, the shortest path (which leads to the fast reach time) to the hospital is calculated. The bat algorithm provides offline data for a possible combination of different source and destination coordinates. The offline data are then trained by utilizing a neural network. The neural network is used for finding the shortest routes between source and destination. The performance evaluation of the BA-CNN algorithm is based on the following metrics: end-to-end delay (EED), throughput, and packet delivery fraction (PDF). This BA-CNN is compared with counterparts, including three different existing methods such as TBM, TVR, and SAODV. The experiments demonstrate that the PDF of our method is 0.90 for 10 malicious nodes, which is higher than in the TBM, TVR, and SAODV.
30 April 2023 in Eastern-European Journal of Enterprise Technologies
The ambulance routing problem is one of the capacitated ambulance routing problem forms. It deals with injuries and their requests for saving. Therefore, the main aim of the ambulance routing problem is to determine the minimum (i.e., optimum) required distances of between: 1) accident places and the ambulance station; 2) the location of the nearest hospital and the accident places. Although of the efforts proposed in the literature, determining the optimum route is crucial. Therefore, this article seeks to tackle ambulance vehicle routing in smart cities using Harris Hawks Optimization (HHO) algorithm. It attempts to take the victims as quickly as possible and confidently. Several engineering optimization problems confirm that HHO outperforms many well-known Swarm intelligence approaches. In our system, let’s use the node approach to produce a city map. Initially, the control station receives accident site information and sends it to the hospital and the ambulance. The HHO vehicle routing algorithm receives data from the driver; the data includes the location of the accident and the node position of the ambulance vehicle. Then, the driver’s shortest route to the accident scene by the HHO. The locations of the accident and hospital are updated by the driver once the car reaches the accident site. The fastest route (which results in the least travel time) to the hospital is then determined. The HHO can provide offline information for a potential combination of the coordinates of destination and source. Extensive simulation experiments demonstrated that the HHO can provide optimal solutions. Furthermore, performance evaluation experiments demonstrated the superiority of the HHO algorithm over its counterparts (SAODV, TVR, and TBM methods). Furthermore, for ten malicious nodes, the PDF of the algorithm was 0.91, which is higher than the counterparts
We present an any-to-one voice conversion (VC) system, using an autoregressive model and LPCNet vocoder, aimed at enhancing the converted speech in terms of naturalness, intelligibility, and speaker similarity. As the name implies, non-parallel any-to-one voice conversion does not require paired source and target speeches and can be employed for arbitrary speech conversion tasks. Recent advancements in neural-based vocoders, such as WaveNet, have improved the efficiency of speech synthesis. However, in practice, we find that the trajectory of some generated waveforms is not consistently smooth, leading to occasional voice errors. To address this issue, we propose to use an autoregressive (AR) conversion model along with the high-fidelity LPCNet vocoder. This combination not only solves the problems of waveform fluidity but also produces more natural and clear speech, with the added capability of real-time speech generation. To precisely represent the linguistic content of a given utterance, we use speaker-independent PPG features (SI-PPG) computed from an automatic speech recognition (ASR) model trained on a multi-speaker corpus. Next, a conversion model maps the SI-PPG to the acoustic representations used as input features for the LPCNet. The proposed autoregressive structure enables our system to produce the following prediction step outputs from the acoustic features predicted in the previous step. We evaluate the effectiveness of our system by performing any-to-one conversion pairs between native English speakers. Experimental results show that the proposed method outperforms state-of-the-art systems, producing higher speech quality and greater speaker similarity.
9 February 2024 in International Journal of Interactive Mobile Technologies (iJIM)
This paper introduces three system architectures for speaker identification that aim to overcome the limitations of diarization and voice-based biometric systems. Diarization systems utilize unsupervised algorithms to segment audio data based on the time boundaries of utterances, but they do not distinguish individual speakers. On the other hand, voice-based biometric systems can only identify individuals in recordings with a single speaker. Identifying speakers in recordings of natural conversations can be challenging, especially when emotional shifts can alter voice characteristics, making gender identification difficult. To address this issue, the proposed architectures include techniques for gender, emotion, and diarization at either the segment or group level. The evaluation of these architectures utilized two speech databases, namely VoxCeleb and RAVDESS (Ryerson audio-visual database of emotional speech and song) datasets. The findings reveal that the proposed approach outperforms the strategy level in terms of recognition results, despite the real-time processing advantage of the latter. The challenge of identifying multiple speakers engaging in a conversation while considering emotional changes that impact speech is effectively addressed by the proposed architectures. The data indicates that the gender and emotion classification of diarization achieves an accuracy of over 98 percent. These results suggest that the proposed speech-based approach can achieve highly accurate speaker identification.
1 October 2023 in TELKOMNIKA (Telecommunication Computing Electronics and Control)