Voice Control Systems Research Articles

How Does Variation in AI Performance Affect Trust in AI-infused Systems: A Case Study With In-Vehicle Voice Control Systems

How Does Variation in AI Performance Affect Trust in AI-infused Systems: A Case Study With In-Vehicle Voice Control Systems

When Do Users Prefer Voice Control Systems in Vehicles? A Survey of Chinese Drivers

Voice control systems (VCSs) are becoming increasingly common in modern vehicles, but concerns are still expressed by drivers about adopting them in smart cockpits. Previous research on privacy and recognition accuracy may not fully address users’ needs with regard to VCSs. In addition, studies on the safety impacts of different interaction modalities may not reveal drivers’ preference for interaction modalities, given that driver choice may not be solely based on interaction safety. Thus, to better understand and optimize VCSs design, an online survey was conducted in China to investigate drivers’ choices between VCSs and manual interaction. In total, we analyzed 168 drivers’ preferred interaction modality (i.e., either VCSs or manual interaction) in different scenarios defined by passenger presence (i.e., present versus absent), traffic complexity (i.e., low versus high traffic density), roadway type (i.e., urban road versus highway), and task characteristics (i.e., six in-vehicle interaction tasks). Compared with manual interactions, drivers preferred VCSs when driving alone, driving on highways, and driving in complex traffic. Further, drivers preferred VCSs when in-vehicle tasks were more distracting and time-demanding (e.g., compiling a text message). The preference for interaction modalities could also be affected by the characteristics of drivers: those with lower driving frequencies and higher VCSs familiarity and those who perceived higher VCSs usability tended to choose VCSs as an interaction modality for in-vehicle interaction tasks. These findings offer insights for adaptive interface design and future optimization of VCSs.

Method for asynchronous analysis of a glottal source based on a two-level autoregressive model of the speech signal

The task of analyzing a glottal source over a short observation interval is considered. The acute problem of insufficient performance of known methods for analyzing a glottal source is pointed out, regardless of the mode of data preparation: synchronous with the main tone of speech sounds or asynchronous. A method for analyzing the glottal source based on a two-level autoregressive model of the speech signal is proposed. Its software implementation based on the high-speed Burg-Levinson computational procedure is described. It does not require synchronization of the sequence of observations used with the main tone of the speech signal and is characterized by a relatively small amount of computational costs. Using the described software implementation, a full-scale experiment was set up and conducted, where the vowel sounds of the control speaker’s speech were used as the object of study. Based on the results of the experiment, the increased performance of the proposed method was confirmed and its requirements for the duration of the speech signal during voice analysis in real time were formulated. It is shown that the optimal duration is in the range from 32 to 128 ms. The results obtained can be used in the development and research of digital speech communication systems, voice control, biometrics, biomedicine and other speech systems where the voice characteristics of the speaker’s speech are of paramount importance.

Laser-Based Command Injection Attacks on Voice-Controlled Microphone Arrays

Voice-controlled (VC) systems, such as mobile phones and smart speakers, enable users to operate smart devices through voice commands. Previous works (e.g., LightCommands) show that attackers can trigger VC systems to respond to various audio commands by injecting light signals. However, LightCommands only discusses attacks on devices with a single microphone, while new devices typically use microphone arrays with sensor fusion technology for better capturing sound from different distances. By replicating LightCommands’s experiments on the new devices, we find that simply extending the light scope (just as they do) to overlap multiple microphone apertures is inadequate to wake up the device with sensor fusion. Adapting LightCommands’s approach to microphone arrays is challenging due to their requirement for multiple sound amplifiers, and each amplifier requires an independent power driver with unique settings. The number of additional devices increases with the microphone aperture count, significantly increasing the complexity of implementing and deploying the attack equipment. With a growing number of devices adopting sensor fusion to distinguish the sound location, it is essential to propose new approaches to adapting the light injection attacks to these new devices. To address these problems, we propose a lightweight microphone array laser injection solution called LCMA (Laser Commands for Microphone Array), which can use a single laser controller to manipulate multiple laser points and simultaneously target all the apertures of a microphone array and input light waves at different frequencies. Our key design is to propose a new PWM (Pulse Width Modulation) based control signal algorithm that can be implemented on a single MCU and directly control multiple lasers via different PWM output channels. Moreover, LCMA can be remotely configured via BLE (Bluetooth Low Energy). These features allow our solution to be deployed on a drone to covertly attack the targets hidden inside the building. Using LCMA, we successfully attack 29 devices. The experiment results show that LCMA is robust on the newest devices such as the iPhone 15, and the control panel of the Tesla Model Y.

Utilizing Near-Ultrasound Inaudible Trojan (NUIT) to compromise microphone security

Voice Control Systems (VCSs) provide a user-friendly interface for issuing voice commands to smart devices. However, the security of VCSs remains a challenge, as evident from two distinct attack classes: (i) inaudible attacks, feasible when the attacker is in proximity to the victim, and (ii) audible attacks, which can be executed remotely by embedding attack signals into audio. In talk introduces a novel class of attacks known as Near-Ultrasound Inaudible Trojan (NUIT). NUIT attacks combine the advantages of both classes, being both inaudible and capable of remote execution. Additionally, NUIT attacks can achieve end-to-end unnoticeability, a crucial aspect often overlooked in the literature. Notably, NUIT attacks leverage victim speakers to compromise victim microphones and associated VCSs, eliminating the need for specialized speakers. The feasibility of NUIT attacks is demonstrated, and an effective defense against them is proposed.

Breaking the silence: Innovation in wake word activation

Trigger term, also referred to as a heat phrase or activation phrase, is a particular expression or wording that activates a voice-controlled system like a virtual aide when spoken. These trigger terms are widely employed in various devices, including mobile phones, intelligent domestic appliances, and automobiles. The precise identification of trigger terms is crucial to guarantee that the voice-controlled system only responds when the trigger term is uttered, disregarding other sounds. Numerous methods, such as regulation-based approaches and machine learning algorithms like support vector machines, concealed Markov models, and profound neural networks, have been proposed for identifying trigger terms. Elements like ambient noise, pronunciation variations, and accents may impact the accuracy of trigger term identification, and researchers are continuously exploring ways to enhance the resilience of these systems. As the utilization of voice-controlled systems continues to grow, the development of reliable and accurate methods for detecting trigger terms is increasingly imperative to ensure the seamless and efficient operation of these systems.

Factors Affecting the Adoption of Artificial Intelligence Voice Control Technology: A Case Study of Xiaomi Smart Devices

The research aims to examine the factors influencing the adoption of artificial intelligence voice control technology. The purposive sampling technique was applied to collect the sample of 400 people who had experience in using Xiaomi smart device. The online questionnaire is the instrument for data collection. Descriptive statistics are used to analyze the respondents’ characteristic. The inferential statistics include Confirmatory Factor Analysis (CFA) and Structural Equation Modeling (SEM) are applied to disclose the impact of utilitarian benefits, hedonic benefits, symbolic benefits, social presence, and social attraction on the adoption of artificial intelligence voice control technology. The results of this research can be used to improve the voice command technology for future users. Additionally, the study’s report can be used for developing voice control systems in smart devices for other business enterprises. Subsequently, the users’ concern about the risk of using voice commands would subside and this technology would be widely accepted.

Expanding functionality of the autonomous voice control system

The article describes the main advantages of voice control systems, notes the shortcomings of the first implemented voice control systems, their functional limitations, and limited use in comparison with modern voice assistants, and also formulates the shortcomings of modern voice control systems. It is noted that with the expansion of functionality, the scope of voice control systems is expanding significantly. The recommendations of leading companies that implement voice technologies, indicating the need to continue research and development in this area. Due to complex calculations in speech recognition, there is a need to use high-speed computers, which makes it difficult to create compact and autonomous voice control systems that work without an Internet connection. At the same time, the growing need for autonomous voice control systems requires a reduction in computational costs, an increase in the speed of calculations by simplifying the methods and algorithms of these systems, implemented in software and hardware in the form of autonomous electronic devices. Methods are distinguished that allow performing recognition with sufficient reliability, such as hidden Markov models and the method for determining mel-frequency cepstral coefficients. In addition, methods are proposed to reduce computational complexity by reducing the number of similar-sounding words included in the command and the commands themselves, and to increase the speed of calculations, the use of electronic components is proposed in the implementation of computational algorithms and their parts. The operability and the possibility of practical application of ways to expand functionality and reduce computational complexity for autonomous voice control systems have been tested and confirmed experimentally, and some of them correspond to the recommendations of leading companies implementing voice technologies.

Voice Recognition Control System Based on Cloud Computing and IoT Sensors

In order to meet the increasing demand for accuracy, stability, and response speed of the voice recognition function of smart home products, the author proposes a smart home voice control system scheme based on the use of smart home Internet gateways to complete remote voice control and environmental monitoring tasks. This method firstly tests the existing home appliance network of the smart home voice control system to verify the feasibility of the author’s design. Then, the voice recognition task is transferred to the cloud server, and the smart home Internet gateway only needs to perform data upload, command execution, and protocol analysis. Then, the test set tests the speech enhancement algorithm in the Aishell library. Experimental results show that the processing speed of the voice enhancement method designed by the author is much higher than that of DNN-noise classification, and its processing speed can reach 0.179 s, which is more suitable as a voice enhancement method for smart home voice control systems. Using this method for voice recognition control can meet the increasing demands of the voice recognition function of smart home products for accuracy, stability, and response speed and make people’s lives more convenient and comfortable.

Voice spoofing detector: A unified anti-spoofing framework

Voice controlled systems (VCS) in Internet of Things (IoT), speaker verification systems, voice-based biometrics, and other voice-assistant-enabled systems are vulnerable to different spoofing attacks i.e., replay, cloning, cloned-replay, etc. VCS are not only susceptible to these attacks in a non-network environment, but they are also vulnerable to multi-order spoofing attacks in networked IoT. Additionally, deepfakes with artificially generated audio pose a great threat to the all systems having voice-interfaces. Most of the existing countermeasures against these voice spoofing attacks work for only one specific attack (e.g. voice replay) and fail to generalize this for other classes of spoofing attacks. Additionally, generalization is also crucial for cross-corpora evaluation. Thus, there exists a need to develop a unified voice anti-spoofing framework capable of detecting multiple spoofing attacks. This work presents a unified anti-spoofing framework that uses novel (ATCoP-GTCC) features to combat the variety of voice spoofing attacks. The proposed novel acoustic-ternary co-occurrence patterns (ATCoP) encode the co-occurrence of similar patterns between the center and neighboring samples. Our experiments demonstrate that ATCoP can better capture the microphone induced distortions in replays, unnatural prosody and algorithmic artifacts in cloned samples, and both the distortions and artifacts in cloned-replays including compression on multi-hop attacks in the spoofing samples. The performance of ATCoP could be further enhanced by the Gammatone cepstral coefficients. To evaluate the effectiveness of the proposed anti-spoofing system for multi-order replay and cloned-replay attacks detection, we created a diverse voice spoofing detection corpus (VSDC) containing multi-order replay and cloned-replay audios against the bonafide and cloned audio recordings, respectively. Experimental results obtained on VSDC, ASVspoof 2019, Google’s LJ Speech, and YouTube deepfakes datasets illustrate the effectiveness of the proposed system in terms of accurate detection for a variety of voice spoofing attacks.

Neural Network Method for Determining The Direction of a Person's Gaze based on a Web Camera Image Analysis

Determining the direction of a person's gaze improves the accuracy of the voice control systems, which are relevant in the actively developing voice assistant creation field. This paper proposes a neural network method for determining the gaze direction based on the web camera image analysis. In the course of the paper, a corpus of data was collected and marked up with preprocessed data from the web camera and the point of view direction on the monitor screen. A neural network model was built based on fully connected and convolutional layers. The created neural network model for determining the gaze direction demonstrated an improvement of 13% in pixel error on the monitor screen compared to the existing open-source solutions. The created neural network model was implemented in the voice control system of a mobile robot, which facilitated minimization of the ambiguity in the analysis of movement commands towards objects.

Analytical Survey : Speech Recognition Methods Used In Voice Recognition Techniques


 
 
 
 This paper mainly studies Speech recognition methods used in voice control systems Analytical survey, Construction of the system voice control is now urgent. Such systems can greatly facilitate user interaction with the computer system.This is especially the idea developed in the concept of Moreover, sometimes the voice Interface is a necessary component of,for example, when it comes to people with disorders of the musculoskeletal system., The work proposed describes general methods of speech recognition based on hidden Markova models (HMM), for in ability to design control system, has high efficiency through voice commands save time and effort and shorten the distance through this circuit get a security system depends on it registered only the person who by scoring and Help people with special needs to rely on themselves in some work
 
 © 2018 JASET, International Scholars and Researchers Association
 
 
 

Research on the oscillation of cooling grills used for terminal devices for voice communication control systems used by air traffic controllers

During this research, the analytical and numerical oscillation models have been deduced of cooling grills of terminal devices for voice communication control systems used by air traffic controllers. The adequacy of obtained models was verified by a full-scale experiment. Vibration stability of telecommunication terminal cases has high importance for the high-quality reproduction of voice information. Obtained results make it possible to solve two problems at once – problem of reducing heat pressure and of increasing vibration stability of audio reproduction device cases.

Accent Recognition with Hybrid Phonetic Features.

The performance of voice-controlled systems is usually influenced by accented speech. To make these systems more robust, frontend accent recognition (AR) technologies have received increased attention in recent years. As accent is a high-level abstract feature that has a profound relationship with language knowledge, AR is more challenging than other language-agnostic audio classification tasks. In this paper, we use an auxiliary automatic speech recognition (ASR) task to extract language-related phonetic features. Furthermore, we propose a hybrid structure that incorporates the embeddings of both a fixed acoustic model and a trainable acoustic model, making the language-related acoustic feature more robust. We conduct several experiments on the AESRC dataset. The results demonstrate that our approach can obtain an 8.02% relative improvement compared with the Transformer baseline, showing the merits of the proposed method.

Digitization of analog signals

We live in the world of digital technologies – everyone has a digital phone, television has switched to a digital broadcasting format as it is more noise-immune, digital processes are literally in every household appliance, from the iron to the computer on which this article was written. Digital technologies simplify our lives, some operations performed by humans require large material costs, for example, writing text on a typewriter and computer vary greatly. The gain of the computer is especially evident when editing the test. They brought us comfort – how nice it is when the processor that monitors the temperature in the house increased the heat supply during a cold snap or reduced it in order to save money in the absence of people in the house.But in order for the digital system to perform this or that action, it needs a command coming from the sensor. It can be a temperature, humidity, pressure sensor. Or maybe a microphone used in voice control systems. All these sensors, without which the operation of a digital system is impossible, give an analog signal that changes its value over time. The digital system is not sensitive to such a signal. It “does not understand” the signal. The problem is solved by ADC (analog-to-digital converters). They have a different structure, varying degrees of complexity, the device, depending on the parameters of the digitized signal. For example, the temperature in the house changes very slowly, even when warming up or when the heating is turned off in frost, the temperature rises or, accordingly, does not fall faster than one degree per hour. The ADC speed requirements for the temperature sensor are very low. Instead, a microphone is used to receive voice commands. In order to distinguish the voices of people and to carry out the commands of some people and not to carry out commands from others, processing of the signal spectrum with a width of kilohertz is required - which means that the signal level will change at a frequency of thousands of times per second. This is a very high demand. The different types of ADCs, their design and application will be discussed in this article.

Denoising and Enhancement Speech Signal Using Wavelet

Speech enhancement aims to improve the quality and intelligibility of speech using various techniques and algorithms. The speech signal is always accompanied by background noise. The speech and communication processing systems must apply effective noise reduction techniques in order to extract the desired speech signal from its corrupted speech signal. In this project we study wavelet and wavelet transform, and the possibility of its employment in the processing and analysis of the speech signal in order to enhance the signal and remove noise of it. We will present different algorithms that depend on the wavelet transform and the mechanism to apply them in order to get rid of noise in the speech, and compare the results of the application of these algorithms with some traditional algorithms that are used to enhance the speech. The basic principles of the wavelike transform are presented as an alternative to the Fourier transform. Or immediate switching of the window The practical results obtained are based on processing a large database dedicated to speech bookmarks polluted with various noises in many SNRs. This article tends to be an extension of practical research to improve speech signal for hearing aid purposes. Also learn about the main frequency of letters and their uses in intelligent systems, such as voice control systems.

Enhancement of Noisy Reverberant Speech Using Polynomial Matrix Eigenvalue Decomposition

Speech enhancement is important for applications such as telecommunications, hearing aids, automatic speech recognition and voice-controlled systems. Enhancement algorithms aim to reduce interfering noise and reverberation while minimizing any speech distortion. In this work for speech enhancement, we propose to use polynomial matrices to model the spatial, spectral and temporal correlations between the speech signals received by a microphone array and polynomial matrix eigenvalue decomposition (PEVD) to decorrelate in space, time and frequency simultaneously. We then propose a blind and unsupervised PEVD-based speech enhancement algorithm. Simulations and informal listening examples involving diverse reverberant and noisy environments have shown that our method can jointly suppress noise and reverberation, thereby achieving speech enhancement without introducing processing artefacts into the enhanced signal.

Integration of Industrially-Oriented Human-Robot Speech Communication and Vision-Based Object Recognition

This paper presents a novel method for integration of industrially-oriented human-robot speech communication and vision-based object recognition. Such integration is necessary to provide context for task-oriented voice commands. Context-based speech communication is easier, the commands are shorter, hence their recognition rate is higher. In recent years, significant research was devoted to integration of speech and gesture recognition. However, little attention was paid to vision-based identification of objects in industrial environment (like workpieces or tools) represented by general terms used in voice commands. There are no reports on any methods facilitating the abovementioned integration. Image and speech recognition systems usually operate on different data structures, describing reality on different levels of abstraction, hence development of context-based voice control systems is a laborious and time-consuming task. The aim of our research was to solve this problem. The core of our method is extension of Voice Command Description (VCD) format describing syntax and semantics of task-oriented commands, as well as its integration with Flexible Editable Contour Templates (FECT) used for classification of contours derived from image recognition systems. To the best of our knowledge, it is the first solution that facilitates development of customized vision-based voice control applications for industrial robots.

Improved hand prostheses control for transradial amputees based on hybrid of voice recognition and electromyography

The control of prostheses and their complexities is one of the greatest challenges limiting wide amputees' use of upper limb prostheses. The main challenges include the difficulty of extracting signals for controlling the prostheses, limited number of degrees of freedom (DoF), and cost-prohibitive for complex controlling systems. In this study, a real-time hybrid control system, based on electromyography (EMG) and voice commands (VC) is designed to render the prosthesis more dexterous with the ability to accomplish amputee's daily activities proficiently. The voice and EMG systems were combined in three proposed hybrid strategies, each strategy had different number of movements depending on the combination protocol between voice and EMG control systems. Furthermore, the designed control system might serve a large number of amputees with different amputation levels, and since it has a reasonable cost and be easy to use. The performance of the proposed control system, based on hybrid strategies, was tested by intact-limbed and amputee participants for controlling the HANDi hand. The results showed that the proposed hybrid control system was robust, feasible, with an accuracy of 94%, 98%, and 99% for Strategies 1, 2, and 3, respectively. It was possible to specify the grip force applied to the prosthetic hand within three gripping forces. The amputees participated in this study preferred combination Strategy 3 where the voice and EMG are working concurrently, with an accuracy of 99%.

Voice-Transfer Attacking on Industrial Voice Control Systems in 5G-Aided IIoT Domain

At present, specific voice control has gradually become an important means for 5G-Internet-of-Things-aided industrial control systems, such as controlling the operation and adjustment of industrial Internet of Things equipment through telephone voice of the controller. However, the security of specific voice control system needs to be improved, because the voice cloning technology based on transfer learning can easily simulate the voice of the controller, which may lead to industrial accidents and other potential security risks. Therefore, this article mainly aims to study and understand the principle of voice cloning attack technology, putting forward a voice clone attack method, in order to prepare for the construction of a specific voice recognition system in the future. At present, the key technology of voice cloning attack is how to solve the problem that the target speaker's personalized speech with high quality cannot be synthesized under small samples. In fact, voice cloning is a very challenging problem because speech is more difficult to be represented in the hidden space of the model. We propose a transductive voice transfer learning method to learn the predictive function from the source domain and fine-tune in the target domain adaptively. The target learning task and the source learning task are both synthesizing speech signals from the given audio, while the datasets of both domains are different. By adding different penalty values to each instances and minimizing the expected risk, an optimal precise model can be learned. In addition, an evaluation method to verify the audio similarity of the target speaker was given to show the similarity between the synthesized audio and the original audio. Many details of the experimental results show that our method can effectively synthesize the speech of the target speaker with small samples.