Recognize Voice Commands Research Articles

Design of voice command recognition chip based on heterogeneous acceleration

Abstract Speech, as one of the earliest forms of communication used by humans, can effectively convey information. However, the current deep neural network models for speech recognition are generally large in scale and can only be deployed in the cloud, which imposes high deployment environment requirements and power consumption, thereby limiting their implementation on embedded devices. In the context of end-to-end speech recognition, a series of challenges are encountered, including power consumption constraints, computing power limitations, network dependencies, privacy protection, bandwidth restrictions, and communication delays. To address these issues, this paper proposes the design of an end-to-end voice command recognition chip based on deep neural networks specifically for recognizing voice commands in specific scenarios. This chip achieves low power consumption and minimal delay in recognition. Additionally, we introduce a weighted, overloadable chip architecture to enable seamless scene migration, ultimately aiming to resolve the aforementioned challenges.

Smart Dispenser Using Voice Recognition as an Assistive Device for the Visually Impaired with a Second-Order IIR Filter Algorithm

This research developed a Smart Dispenser with Voice Recognition technology as a tool to help the visually impaired in taking drinking water. The system is designed using a Voice Recognition sensor, an ISD 1820 sensor, and a microcontroller to enable the operation of the dispenser via voice commands. The test was carried out to analyze the performance of the system in recognizing voice commands and removing water. The results showed a voice detection accuracy rate of 28%, with factors such as environmental noise and sound intonation affecting the effectiveness of the system. The study also identified several areas of improvement, including the use of a DC water pump to increase the flow rate, the addition of an automatic feature to stop the flow of water when a glass is taken, and the use of a DC fan to maintain the sensor's optimal temperature. Further development is needed to improve the accuracy and efficiency of the system so that it can more effectively help the visually impaired in daily life.

VOICE CONTROL CAR USING MACHINE LEARNING

The "Voice-Controlled Car Using Machine Learning" project aims to create an innovative and user- friendly vehicle control system by integrating machine learning technology. This system provides a hands-free driving experience, making it safer and more convenient for drivers. The project involves developing a smart car prototype equipped with various sensors and microcontrollers. Machine learning algorithms are implemented to recognize voice commands and translate them into specific actions for the vehicle, such as accelerating, braking, turning, and even parking. The core components of the system include a speech recognition module, a control interface, and a real-time data processing unit. The project not only demonstrates the potential of machine learning in the automotive industry but also promotes a more accessible and user-centric approach to vehicular control. Keywords: “Speech Recognition”, command interpretation”, audio processing”, Wake word Detection”.

Improved Speaker Recognition System Using Automatic Lip Recognition

The paper is focused on the relevant problem of speech recognition using additional sources besides the voice itself, in conditions in which the quality or availability of audio information is inadequate (for example, in the presence of noise or additional speakers). This is achieved by using automatic lip recognition (ARL) methods, which rely on non-acoustic biosignals generated by the human body during speech production. Among the applications of this approach are medical applications, as well as processing voice commands in languages with poor audio conditions. The aim of this work is to create a system for speech recognition based on a combination of speaker lip recognition (SSI) and context prediction. To achieve this goal, the following tasks were performed: to substantiate the systems for recognizing voice commands of a silent voice interface (SSI) based on a combination of two neural network architectures, to implement a model for recognizing visemes based on the CNN neural network architecture and an encoder-decoder architecture for the LSTM neural recurrent network model for analyzing and predicting the context of a speaker’s speech. The developed system was tested on a chosen dataset. The results show that the recognition error in different conditions averages from 4,34% to 5,12% for CER and from 5,52% to 6,06% for WER for the proposed ALR system in 7 experiments, which is an advantage over the LipNet project, which additionally processes audio data for the original without noise.

A multi-task network for speaker and command recognition in industrial environments

In industrial environments, it is crucial to establish a strong collaboration between humans and robots to enhance productivity. However, the nature of the work demands that workers have the authority to provide specific instructions to the robots. The scientific community has extensively investigated these dual requirements, aiming to develop advanced systems capable of recognizing voice commands and implementing speaker authentication. Nevertheless, in the industrial context, these tasks should be executed simultaneously on low-cost and low-power embedded devices that can be mounted on board the robotic platform. To overcome this challenge, we propose a multi-task network for Speech-Command Recognition and Speaker Identification. Additionally, we employ the GradNorm adaptive algorithm to address the issue of task imbalance. To evaluate the proposed system, we introduce a new dataset, MIVIA-ISC, consisting of 20,857 samples uttered by 562 speakers for 31 distinct commands. Our approach significantly reduces the network size by 47% and its execution time by 48% compared to the commonly used methodology, which employs one network for each task. Furthermore, our approach demonstrates a significant improvement in the accuracy of the Speaker Identification task, achieving an 11% increase compared to the corresponding single-task network. Importantly, this enhancement is achieved without compromising the accuracy of the Speech-Command Recognition task, which experiences only a minimal 3% decrease in performance.

VOICE ASSISTANT FOR FINDING ITEMS IN CLOTHING STORE

The paper presents the use of machine learning in audio file processing, the use of neural networks to recognize voice commands, and the development of a voice assistant for finding products in a clothing store. 
 The purpose of the work is to develop a voice assistant for finding items in a clothing store.
 As a result, a dataset containing 30 categories and 3095 audio recordings was created, a neural network model was trained using the collected data, and an accuracy of 96.02% was achieved; WER: 0.0398; CER: 0.0087. The model was integrated with the search system into the voice assistant API, which allows you to record from a microphone, convert audio to text, break the text into keywords, and search the database using the obtained keywords. The speech recognition system has shown stable and high accuracy when recording from a microphone. This provides users with reliable and accurate recognition results even when using simple microphones. Search allows you to find results by keywords and names of items, and there is a function to recommend similar items if nothing was found at the user's request. The system's flexibility allows it to understand language contexts and does not depend on the order of words in a phrase, as the search is performed by keywords separately. The research analyzes the literature and compares existing approaches to voice command recognition. The methods of audio signal processing are described. The problem of searching for items in a clothing store was analyzed, and the current state of the market and demand for the technology were investigated. The practical value is that the developed voice assistant is specialized and optimized for searching for goods in a clothing store, allows solving the task of finding goods according to specified criteria, and simplifies the search task for different groups of people.

A Smart Metasurface for Electromagnetic Manipulation Based on Speech Recognition

In this work, we propose and realize a smart metasurface for programming electromagnetic (EM) manipulations based on human speech recognition. The smart metasurface platform is composed of a digital coding metasurface, a speech-recognition module, a single-chip computer, and a digital-to-analog converter (DAC) circuit, and can control EM waves according to pre-stored voice commands in a smart way. The constructed digital metasurface contains 6 × 6 super unit cells, each of which consists of 4 × 4 active elements with embedded varactor diodes. Together with the DAC and single-chip computer, the speech-recognition module can recognize voice commands and generate corresponding voltage sequences to control the metasurface. In addition, a genetic algorithm (GA) is adopted in the design of the metasurface for efficiently optimizing the phase distributions. To verify the performance of the smart metasurface platform, three typical functions are demonstrated: radar cross-section reduction, vortex beam generation, and beam splitting. The proposed strategy may offer a new avenue for controlling EM waves and establishing a link between EM and acoustic communications.

Исследование статистики развития нейронных сетей и их роли в сфере видеопроизводств

The development of technology these days is amazing in its rapidity. Computer and network technologies have come a long way from local file sharing services to a global repository that absorbs the knowledge of many generations of humanity. However, now a new milestone in technology development is gaining momentum - artificial neural networks, which are something different than the prototypes of artificial intelligence shown to us in popular culture - films, TV series and books. These are assistants that are developing every day, capable of performing a wide range of tasks assigned by a person. Neural networks are becoming more and more rooted in our lives, and we use them without even realizing that they are branches of artificial intelligence. The list of them is quite large, starting from voice assistants aimed at quickly searching for information on the Internet. Due to competition in the technology market, almost every large IT company owns its own voice assistant, which is distributed on devices produced by the companies - personal computers, mobile phones and other smart devices. Among the most influential, it is worth noting the voice assistants ―Alice‖, developed by Yandex, ―Cortana‖ - developed by Microsoft, and ―Siri‖ - a voice assistant for devices created by Apple. Also, in addition to voice assistants that recognize voice commands, there are neural network generators that have found their niche in various branches of modern Internet media, from generating text to creating original images based on user requests. Video content is one of the fundamental pillars of media culture around the world and to this day, the majority of the population finds it preferable due to receiving more information in less time. However, now in view of the development of neural network technologies, the question arises: will actively learning neural networks replace full-fledged video production workers? The article will conduct a study of statistics on the use of neural networks for working with media content, and will also answer the question posed above.

Voice-activated wheelchair: An affordable solution for individuals with physical disabilities

The Low-Cost Voice Controlled Wheelchair with Raspberry Pi is an innovative assistive technology designed to improve the mobility and independence of people with disabilities. This research aims to develop a wheelchair system that can be operated using voice commands at an affordable price, making it accessible to a wider range of individuals with limited mobility. The device is built on the Raspberry Pi, a reasonably priced, credit-card-sized computer, and uses an easy-to-use yet efficient voice recognition technique to let users control the wheelchair with their vocal commands. A Raspberry Pi, a microphone, and motor controllers are some of the system's hardware components. The software uses Python programming language and open-source voice recognition technology to recognize voice commands, making it easy for users to navigate their environment independently. The system has been tested on a prototype and has shown promising results in terms of accuracy and reliability. The Low-Cost Voice Controlled Wheelchair with Raspberry Pi can give disabled persons new levels of mobility and independence, enhancing their quality of life and enhancing their capacity to carry out daily tasks.

Advanced Multimodal Interfaces Design Using Speech Control

The paper discusses methods of developing human-machine interfaces for air-craft cabins using voice control. The use of voice control is relevant for promising multimodal aviation interfaces. The features of methods for recognizing voice commands are considered, as well as system design of human-machine interfaces based on methods for analysing crew tasks.

MULTILEVEL CONTROL OF A TRANSPORT ROBOT

The control systems of modern robots are multilevel. The upper level control of the transport robot by voice commands and simulation of its motion in the software product Microsoft Robotics Developer Studio are considered. Visual Programming Language for creating and debugging applications was used to simulate the motion of the transport robot. An Android tablet or mobile phone with an application is used to recognize voice commands and the robot can be controlled using the Bluetooth interface. A program has been developed that will allow a human to control the robot using speech. An example of simulation of the robot and its motion path using voice control is given. The developed control system allows the transport robot to follow from one target point to another using voice control. At a low level, optimal asynchronous motor based actuator control using a model oriented approach in SimInTech is considered.

Recognition of Voice Commands Based on Neural Network

The paper considers features of voice commands pronunciation models, namely, dependence: system accuracy on number of states for phonemes; system accuracy on learning rate; accuracy of system on value of training set. A speech recognition system based on neural networks is proposed. A speech recognition system is not easy to implement and requires an understanding of speech recognition basics. The developed system is compared with Speech Recognition from Google and Pocket Sphinx. The proposed system can recognize voice commands with an accuracy of 84.4 %.

Remote Control using Voice Recognition based on Arduino

Home automation becomes important, because it gives the user convenient and easy method to use home appliances. This paper aims to help people with special needs or physical disabilities and injuries by paralysis to control any device using infrared technology using voice commands based on the voice recognition system (voice recognition unit V3) system can recognize voice commands, convert them to desired data coordination and data transmission via IR transmitter and microcontroller (Arduino Uno) Receiving this signal by IR sensor to control TV receiver then get a full remote control that works by voice commands. The software consists of a Micro C language programmable microcontroller. This system is of low cost and flexible with growing variety of devices that can be controlled.

Voice recognition system for controlling electrical appliances in smart hospital room

Nowadays, most hospitals have new problem that is lack of medical nurse due to the number of patient increas rapidly. The patient especially with physical disabilities are difficult to control the switch on electrical appliances in patient’s room. This research aims to develope voice recognition based home automation and being applied to patient room. A miniature of patient’s room are made to simulate this system. The patient's voice is received by the microphone and placed close to the patient to reduce the noise.V3 Voice recognition module is used to voice recognition process. Electrical bed of patient is represented by mini bed with utilising motor servo. The lighting of patient room is represented by small lamp with relay. And the help button to call the medical nurse is represented by buzzer. Arduino Uno is used to handle the controlling process. Six basic words with one syllable are used to command for this system. This system can be used after the patient's voice is recorded. This system can recognize voice commands with an accuracy 75%. The accuracy can be improved up to 85% by changing the voice command into two syllables with variations of vowels and identical intonation. Higher accuracy up to 95% can be reached by record all the subject’s voice.

Comparative analysis of speech recognition algorithms in UAV voice control system

The article proposes to perform a comparative analysis of the presented algorithms for processing voice control signals for an unmanned aerial vehicle, which can be implemented on processors with low computing power using online processing in real time. It is shown that these approaches are effective in improving the accuracy of speech recognition in the presence of various types of noise and a sound-reflecting control environment, which is an important problem in voice control systems for an unmanned aerial vehicle. An algorithm for calculating the mel-frequency cepstral coefficients, which appear in the role of the main features of speech recognition, is presented. A comparative analysis of two methods of distinguishing informative features of speech recognition in the voice control system of an unmanned aerial vehicle was made, namely, mel-frequency cepstral factors and the coefficients obtained with the aid of a linear prediction algorithm, where as a result of the conducted scientific experiment, under the influence of given noise, it was concluded that in these problems, the optimal method of exclusion is the mel-frequency cepstral factors, since they show the best value for obsalutnomu criterion of speech recognition quality. The expediency of using the proposed system for recognizing voice commands of an unmanned aerial vehicle based on the cepstral analysis is substantiated and experimentally proved. The obtained results of the experimental research allow to draw a conclusion about the advisability of further practical application of the developed system for recognizing voice commands for the control of an unmanned aerial vehicle on the basis of a cepstral analysis.

Transport Robot Control by Voice Commands

В системах управления современных роботов используется нечеткая логика, поэтому целесообразно подавать команды управления роботом на основе нечеткой логики, которая позволит роботам эффективно выполнять поставленные задачи. В статье рассматривается управление транспортным роботом голосовыми командами и моделирование его движения в программном продукте Microsoft Robotics Developer Studio. Для моделирование движения транспортного робота использовался язык Visual Programming Language (VPL) - среда визуального программирования для создания и отладки приложений. Для распознавания голосовых команд используется планшетный компьютер или мобильный телефон с операционной системы Android с приложением, при помощи которого можно управлять роботом с использованием интерфейса Bluetooth. Разработана программа, которая позволит человеку управлять роботом с помощью речи. Получена визуальная программа управления движением транспортного робота. Приведен пример моделирования робота и его траектория движения с использованием голосового управления. Разработанная система управления позволяет следовать транспортному роботу от одной целевой точки к другой с использованием голосового управления. Задача управления роботом с использованием голосового управления существенно упрощается, поскольку она практически не требует специальных навыков от оператора. В конечном счете, голосовое управление облегчит использование роботов в промышленности, быту и других областях. Управлять можно будет не только роботами, но и другими устройствами, имеющими микропроцессорное управление.

Cross-correlation portraits of voice signals in the problem of recognizing voice commands according to patterns

A way for recognizing voice commands (VCs) in the noises with a probability of proper recognition higher than 92% and a signal/noise ratio of 1---6 dB, if the library of pattern voice commands has been generated directly before recognition, is presented in [1]. This method is based on transformation of voice signals into a 2D image: autocorrelation portrait (ACP). The results become significantly worse if the library is prepared long before the recognition, and this is a disadvantage of this method. In this paper we describe the procedure for generating another type of voice command image, which eliminates (to a considerable degree) this disadvantage.

Healthcare Robot Technology Development

The interest in a healthy and happy life for seniors has increased of late as the quality of life in old age has become a hot issue. Moreover, the interest in health has led industrial companies to concentrate on healthcare technology. Healthcare robot technology is one of the biggest issues in the robotics field, and many companies and institutes have sought to develop a healthcare robot technology. The healthcare robot must have certain abilities that would allow it to help a person mentally and/or physically. There is no common interface module, however, for healthcare robot platforms, and few robots that have healthcare abilities exist. Therefore, developing a common interface technology and applying the technology to robot platforms can pave the way for the development of a technology that could improve the physical and mental health of humans. This paper introduces the authors’ research on a common interface module and a healthcare service robot platform. First, a common interface module is one that can be used in common in the healthcare robot area. The research on a common interface module contains the following technologies: biosignal processing modules; voice signal processing modules, which could recognize voice commands; control modules for robot actuators; and a human-friendly design for a healthcare system. Lastly, the research on a healthcare service robot platform contains three types of robot platforms: ChairBot for relaxation, RideBot for exercising, and LifecareBot for human emotions. Each robot platform has common interface modules for healthcare services and can provide specialized intelligent services based on their own objectives. In conclusion, the intellectual healthcare robot is expected to promote a new field and to pave the way for a new product.

Speech/gesture interface to a visual-computing environment

We developed a speech/gesture interface that uses visual hand-gesture analysis and speech recognition to control a 3D display in VMD, a virtual environment for structural biology. The reason we used a particular virtual environment context was to set the necessary constraints to make our analysis robust and to develop a command language that optimally combines speech and gesture inputs. Our interface uses: automatic speech recognition (ASR), aided by a microphone, to recognize voice commands; two strategically positioned cameras to detect hand gestures; and automatic gesture recognition (AGR), a set of computer vision techniques to interpret those hand gestures. The computer vision algorithms can extract the user's hand from the background, detect different finger positions, and distinguish meaningful gestures from unintentional hand movements. Our main goal was to simplify model manipulation and rendering to make biomolecular modeling more playful. Researchers can explore variations of their model and concentrate on biomolecular aspects of their task without undue distraction by computational aspects. They can view simulations of molecular dynamics, play with different combinations of molecular structures, and better understand the molecules' important properties. A potential benefit, for example, might be reducing the time to discover new compounds for new drugs.