Voice Recognizer Research Articles

Boosting Sensing Performance of Flexible Piezoelectric Pressure Sensors by Sb Nanosheets and BaTiO3 Nanoparticles Co‐Doping in P(VDF‐TrFE) Nanofibers Mat

AbstractFlexible piezoelectric pressure sensors (FPPS) received a lot of attention because of their prominent performance in body motion detection and energy recovery. However, it has faced a huge challenge in manufacturing FPPS with high sensitivity, low cost, and high flexibility. Herein, a ternary flexible nanofiber composite with Sb nanosheets and BaTiO3 nanoparticles as the fillers of P(VDF‐TrFE) is prepared by electrospinning. An impressive open‐circuit voltage of 17.1 V and short‐circuit current of 4.4 µA is obtained based on the composite nanofiber mat at a pressure of 128 kPa with 2 Hz frequency. The FPPS not only exhibit a high sensitivity of 96 mV kPa−1 but also an ultrafast response time of 2 ms. Excellent flexibility and reliability are demonstrated by the unchanged open‐circuit voltage after 2400 cycles. The enhanced FPPS performance is owed to the interfacial polarization effect at the inorganic filler‐organic matrix interface, which also enhances the ferroelectric and dielectric properties. The present FPPS is further confirmed to be a real‐time motion monitor and voice recognizer, which can distinguish various actions and sounds according to distinct output voltage signals and is expected to be used in virtual reality devices, robotic electronic skin, haptic simulation, artificial throat, etc.

Addressing the selection bias in voice assistance: training voice assistance model in python with equal data selection

In recent times, voice assistants have become a part of our day-to-day lives, allowing information retrieval by voice synthesis, voice recognition, and natural language processing. These voice assistants can be found in many modern-day devices such as Apple, Amazon, Google, and Samsung. This project is primarily focused on Virtual Assistance in Natural Language Processing. Natural Language Processing is a form of AI that helps machines understand people and create feedback loops. This project will use deep learning to create a Voice Recognizer and use Commonvoice and data collected from the local community for model training using Google Colaboratory. After recognizing a command, the AI assistant will be able to perform the most suitable actions and then give a response. The motivation for this project comes from the race and gender bias that exists in many virtual assistants. The computer industry is primarily dominated by the male gender, and because of this, many of the products produced do not regard women. This bias has an impact on natural language processing. This project will be utilizing various open-source projects to implement machine learning algorithms and train the assistant algorithm to recognize different types of voices, accents, and dialects. Through this project, the goal to use voice data from underrepresented groups to build a voice assistant that can recognize voices regardless of gender, race, or accent. Increasing the representation of women in the computer industry is important for the future of the industry. By representing women in the initial study of voice assistants, it can be shown that females play a vital role in the development of this technology. In line with related work, this project will use first-hand data from the college population and middle-aged adults to train voice assistant to combat gender bias.

Vocal Airmail Service for Visually Impaired through Multimedia Approach

With the advent of technologies in computer, many technological solutions have been implemented for visually impaired so that they can utilize them, and get benefited by them. Considering it as a key idea we have built a desktop application that will help blind people to send and read emails as ordinary people do. In this research paper we describe the VMAIL system architecture for Python platform that can be used by a Blind Person to access e-mails easily. The application uses ‘text to speech’ and voice recognizer to facilitate sending, reading, emails using a computer. Internet has become one of the basic amenities for day-to-day living. Every human being is widely accessing the knowledge and information through internet. However, blind people face difficulties in accessing these text materials, also in using any service provided through internet. The advancement in computer based accessible systems has opened up many avenues for the visually impaired across the globe in a wide way. Audio feedback based virtual environment like, the screen readers have helped Blind people to access internet applications immensely. We describe the Voicemail system architecture that can be used by a Blind person to access e-Mails easily and efficiently

『길 위 1번지』, AI 제임스의 소설:「소설의 기술」과 인공신경망 알고리즘의 글쓰기

This article explores the implications of 1 the Road (2018), a fiction produced by an artificial neural network, for the debate on machine creativity. Unlike most of the recent deeplearning writing programs, its creator Ross Goodwin traveled with this neural network machine from New York to New Orleans in a Cadillac equipped with a surveillance camera, a voice recognizer, and GPS. The AI thus had continuous feeds of fresh images and sounds, not unlike a novelist who hits the road searching for an experience. I argue that we can better understand this attempt of 1 the Road when discussed alongside Henry James’s realist notion of fiction in his essay “The Art of Fiction” (1884). Even though James is often considered an arch-humanist in his belief that fiction is an expression of the unique personality of a novelist, his idea of experience offers a way of understanding Goodwin’s machine as a creative writer. Its algorithmic operation―processing the new sense data via the CNN algorithm and producing the result as writing via the RNN algorithm―is strikingly similar to the workings of the novelist’s mind in James’s view. I do not argue that those sense data are unmediated, and this fact is reflected in the CNN’s algorithm that imitates the human mind. In fact, James asserts that sensory data are the result of mediated selection, which can only be seen through the final result, the writing itself. What I propose is that regardless of whether the machine has a so-called “consciousness,” its writing can affect the reader in similar ways as does a text written by a human being. As long as it works as a linguistic medium to relate the “experience,” which is the algorithmic process of selection and transcription, it is a creative writer.

Virtual safety device for women security

The Crime against the women in our country today is becoming a daily routine. There continues to be increased demand in the development of safety devices to ensure a secure and safe environment for the women. The key objectives are to develop a kind of “Virtual Friendly” machine to protect you from serious troubles. These suggested systems are made up of the GPS, the GSM transmitter, the Microcontrollers (LPC812), Voice Recognizer, a RF transceiver and Temperature Sensors. GPS Receiver receives satellite locations information’s I n the indirect form of the longitude and the latitude. It information is handled by the Microcontroller and this gathered information the sent to the main control point of POLICE. These special temperature sensors continuously sense the body temperature. In a case of a very low temperature indicating some emergency conditions, a kind of SMS alert would be automatically sent. The device can be triggered by the consumer on its own when self-defense is necessary. To ensure maximum performance, functionality such as warning tone, voice recognition is applied to the device. Using the SAVE ME and Support ME commands, for the voice recognizers this circuit is utilized to transmit voice commands. The device is designed as a small package and is easy to operate, so that a simple women can conveniently take these along everywhere.

Virtual Friendly Device for Women Security

Crimes against women is becoming a daily routine in our country, nowadays. An increased demand persists in developing safety devices for ensuring safe and secure environment for women. The main aim is to develop a “Virtual Friendly” device, to safeguard themselves while in trouble. The proposed system developed consists of GPS, GSM modem, Microcontroller (LPC812), RF transceiver, Temperature Sensor, Voice Recognizer. GPS Receiver gets the location information from satellites in the form of latitude and longitude. The Microcontroller processes this information and this processed information is sent to the POLICE control room. The temperature sensor senses the body temperature continuously. SMS alert will be sent in case of low temperature which indicates emergency condition. The system can be activated by the user by themselves, when they are in need of self-defense. Features such as Alarm sound, voice recognition are added to the system to ensure optimum efficiency. The voice recognizer circuit is used to send command with voice by using the codes SAVE ME and HELP ME. The device is designed as a portable kit and easily operated, so that any women can easily carry with them anywhere and everywhere.

임베디드 시스템에서 사용 가능한 적응형 MFCC와 Deep Learning 기반의 음성인식

This paper proposes a noble voice recognition method based on an adaptive MFCC and deep learning for embedded systems. To enhance the recognition ratio of the proposed voice recognizer, ambient noise mixed into the voice signal has to be eliminated. However, noise filtering processes, which may damage voice data, diminishes the recognition ratio. In this paper, a filter has been designed for the frequency range within a voice signal, and imposed weights are used to reduce data deterioration. In addition, a deep learning algorithm, which does not require a database in the recognition algorithm, has been adapted for embedded systems, which inherently require small amounts of memory. The experimental results suggest that the proposed deep learning algorithm and HMM voice recognizer, utilizing the proposed adaptive MFCC algorithm, perform better than conventional MFCC algorithms in its recognition ratio within a noisy environment.

Establishment of Confidence Thresholds for Interactive Voice Response Systems Using ROC Analysis

INTRODUCTION An Interactive Voice Response (IVR) System is a platform for man-machine interaction by the use of voice or keypad. Examples abound. Whenever one calls most large organizations, their initial encounter is with a machine that will prompt the caller for their intent. Usually, such machines will give you options to choose from (Directed Dialog), or it may ask for your input (Open Dialog). In the case of Open Dialog, there is the risk that the machine does not understand a caller input. This is an area where a lot of investigation takes place to deduce why this is the case. The technology for recognizing keyed input is not as challenging as speech technology because each key on the keypad corresponds to a specific sound frequency that cannot be confounded with another key. This technology is called Dual Tone Multi Frequency (DTMF.(i)); and it is a mature technology due to the fact that there is little or no variability in the tone emitted by a particular key. This is not the case with speech. In the case of speech technology, there are several variables that come into play. These include whether a caller barges-into a prompt, whether there is a lot of background noise that may be of similar frequency as the spoken utterance, whether the user is using a cell phone, a speaker phone, or a computer. These, and several other factors, affect the way an IVR system recognizes the caller input. This paper is an attempt to establish guidelines for determining the best settings under which an IVR system should accept a caller input using ROC analysis. REVIEW OF LITERATURE Receiver Operating Characteristics (ROC) analysis has been used in medical imaging to measure diagnostic accuracy (Metz, 2008; Pepe, 2000; Griner, Mayewski, Mushlin, & Greenland, 1981). To diagnose diseases, (McClish, 1989) used this technique to analyze the accuracy of the diagnosis. He preferred this technique because it provided the investigator with all possible combinations of sensitivity and specificity. ROC analysis has been used in the field of radiology (Metz & Obuchowski, 2003). ROC analysis was applied to biomedical informatics, (Lasko, Bhagwat, Zou, & Ohno-Machado, 2005; Brown & Davis, 2006; Hand, & Till, 2001), Signal Detection Theory (Green & Swets, 1966); it provides a precise language and graphic notation for analyzing decision-making in the presence of uncertainty. ROC curves are used extensively in epidemiology and medical research and are frequently mentioned in conjunction with evidence- based medicine (Zweig & Campbell, 1993). Bond and DePaulo (2006) used ROC analysis to study the accuracy of Deception judgments by studying over 20,000 judgments, and came to the conclusion that such analysis correlated strongly with other methods of analysis. In the field of Artificial Intelligence (Fogarty, Baker, & Hudson, 2005), ROC curves have proved useful for the evaluation of machine learning techniques (Flach, 2004; Fawcett, 2006). The approach used in this paper is to extend the use of ROC analysis to Speech Recognition. If an utterance is clearly understood (with high/medium confidence) the caller will be led further down the rest of the call flow. If, however, the IVR engine is not certain what the caller input is, it would be compelled to re-prompt the caller so as to confirm that the original intent was correctly identified. After the second attempt at recognition, for caller inputs that are still not clearly understood by the IVR engine, the caller will be transferred to a live agent. This is what the IVR engine is designed for - to minimize (and possibly eliminate) the cost of transferring to a live agent. THE ENVIRONMENT The Interactive Voice Response (IVR) environment consists of a platform for collecting and analyzing caller utterances using a voice recognizer. The quality of the categorization varies with the parameter settings of the recognizer. The two main parameters of the recognizer are: the energy floor and the confidence threshold. …

Human Tracking Methods Comparison for Smart House Method Comparison in Human Tracking for Smart House Purpose

Smarthouse systems require some information of human locations in the system environment, different smart house system offers different human tracking method. The idea of this paper is to determine which method is better in which condition. This paper discusses amongst RFID, voice recognizer and temperaturesensor. As a result, each method have its own advantages and disadvantages.
 Index Terms - smart house, human tracking, voice recognition RFID, Sensor.

Automatic Food Documentation and Volume Computation Using Digital Imaging and Electronic Transmission

Capturing accurate food intake data from participants enrolled in nutrition studies is essential for understanding relationships between diet and chronic disease (1). Numerous methods are employed to assess dietary intake such as food records, 24-hour recalls, or food frequency questionnaires. While each of these techniques is valuable, the error associated with each is unique. The food record requires a motivated participant, is tedious for some, places attention on the act of eating thus altering intake and is difficult for subjects with low literacy skills (2). Interviewing subjects about the previous day’s intake avoids the reactivity involved when recording current intake, but also requires the individual reporting intake to have good recall skills, knowledge of food names and ability to estimate amounts eaten; and requires a well-trained interviewer which makes this a costly process (2, 3). Food frequency questionnaires are limited by food lists and lack of detail regarding food preparation, and require respondents to summarize past intake over many months or the past year. Such instruments are known to contain significant measurement error (4). While all these methods provide valuable information about dietary intake, improving methodology even modestly would advance our knowledge about the influence of food intake on health. FIVR (Food Intake Visual and voice Recognizer), a subproject of the Genes, Environment, and Health Initiative from the National Institutes of Health (RFA-CA-07-032 at www.gei.nih.gov/index.asp), is designed to use new digital photographing technology to reduce measurement error associated with a food record. The intent is to create a tool that would both increase accuracy of intake records and reduce the recording burden for respondents. Using a mobile phone with a camera (Figure 1), the participant will photograph foods both before and after eating. In this way initial portion size is recorded as well as portions left uneaten. The photographs would be used to identify both the types and amounts of foods consumed. This paper briefly describes the technology and techniques involved. Figure 1 Typical Mobile Phone Interface showing (a). operator instruction screen, (b) menu of activities available and (c) camera poised to record meal. Creating sufficiently detailed images Capturing images of meals using a mobile phone presents its own unique challenges. Identifying foods from a picture requires a clear image; the automatic calculation of the amount eaten (volume) requires three or more clear images to be taken by the mobile phone user. Since a single image will not support estimation of food volume, rather 3-dimensional objects must be viewed at more than one angle (5, 6). The three images in Figure 2 are captured from 3 slightly different angles. A calibration object is also required in the images for determination of 3-dimensional size (see Figure 2). The calibration object (fiduciary marker) included in the images in Figure 2 is a card with black and white squares of known size. However, a standard credit card can be used to establish the relationship between size in image pixels and actual size of the object in milliliters. Images are also required before and after the meal is eaten to document the volume of food consumed. Figure 2 Three images captured by moving the camera using the FIVR mobile phone system. Quality of the image hinges on several factors including resolution (roughly indicated by number of pixels per image). Higher resolution (more pixels per image) creates larger files, which makes transferring images slower and more subject to failure, thus testing and refinement of the image details is integral to developing a successful system. Camera focus is critical since the best volumetric estimation is obtained when the three images are in focus and taken with the plate at the same distance from the camera. With fixed focus cameras, the images will be blurred if not taken at the right distance (which is often too great). With auto-focus cameras, the focusing is assured but the distance still must be maintained by the user. Ways to adjust the image to correct for small variations in distance are still being explored.

1A2-M11 パラボラ集音器を用いた音声認識装置による電動車いす制御

1A2-M11 パラボラ集音器を用いた音声認識装置による電動車いす制御

Using Language Technology to Increase Efficiency and Safety in ATC Communication

Voice communication is a volatile part ofAir Traffic Control (ATC) and thus the InternationalCivilAviationOrganization(ICAO)putsgreatemphasisonimprovingcommunication in ATC. Research has shown that miscommunication happens on average once every hour per radio frequency where there is frequent communication, such as in Terminal Radar Approach Control (TRACON). This paper proposes that a Language Technology System (LTS) can make communication between controller and pilot more reliable and efficient, thus improving safety in aviation. An LTS can for example detect read back errors. It can also directly feed data from the voice recognizer into a Flight Data Processing System or interact with it otherwise as we show. By interviewing air traffic controllers and studying the literature, we identified these and several other examples of use of language technology in ATC. In this paper, we explore one of the identified examples, a system that takes over the communication between controller and pilot in oceanicATC. This system is not meant to control the airspace autonomously, but only relay information. Latest advances in language technology have enabled the development of such a system. The functionality of the proposed LTS is described using scenarios and sequence diagrams.A demonstration LTS using Hex Technology was implemented. A usability test was administered to seven controllers. Their attitude towards the agent was positive and indicates that there is reason for further research. The performance and error logs of the LTS were analyzed and give guidance on further development of a fully functioning LTS for ATC.

User interface for telematics systems

A combination of manual and voice input for a telematics system. The telematics system uses a menu-based interface to communicate with the driver. This menu-based interface is accessible to the driver using both manual interactions and by voice. A reliable manual input mechanism is available using turn or push bottons in addition to voice recognition, which may not be accurate in all operating conditions. Voice recognition inputs can be limited to small numbers of words or phrases that can be more easily distinguished by a voice recognizer. Futhermore, voice outputs of the system are tailored to that they are readily recognizable to the driver. Manual input is optionally rejected in operating conditions in which such input is dangerous or prohibited.

Load-adjusted speech recogintion

A speech recognition system includes a user interface configured to provide signals indicative of a user's speech. A speech recognizer of the system includes a processor configured to use the signals from the user interface to perform speech recognition operations to attempt to recognize speech indicated by the signals. A control mechanism is coupled to the voice recognizer and is configured to affect processor usage for speech recognition operations in accordance with a loading of the processor.

Voice announcement management system

A central control and management station (10) is provided for automatically creating, loading, modifying and/or checking recorded audio announcements at a plurality of remotely located playback devices (20) of a plurality of vendors in a telecommunications network. The station (10) includes a library of voice files (13), an access database (18) containing machine identification details and access instructions for each of the remotely located playback devices (20), a management unit (11) for receiving requests from vendors to load or modify recorded announcements in the playback devices (20) and a control unit (12) which sends the recorded announcements via loading interfaces (17) to the playback devices (20). The station (10) includes or is connected to a text to speech converter (15) and/or a digital voice recording station (30) for creating new announcements so that if a requested announcement is not present in the voice files (13) the management unit (11) can automatically request the creation of a new announcement. The station (10) also includes a voice recognizer (19) for verifying that a recorded announcement has been correctly loaded in a playback device (20).

Speech recognition method and system using compressed speech data

A vocoder based voice recognizer recognizes a spoken word using linear prediction coding (LPC) based, vocoder data without completely reconstructing the voice data. The recognizer generates at least one energy estimate per frame of the vocoder data and searches for word boundaries in the vocoder data using the associated energy estimates. If a word is found, the LPC word parameters are extracted from the vocoder data associated with the word and recognition features are calculated from the extracted LPC word parameters. Finally, the recognition features are matched with previously stored recognition features of other words, thereby to recognize the spoken word.

Keep talking—Performance effectiveness with continuous voice recognition for spreadsheet users

The performance of spreadsheet users was compared for two modes of input to the computer—keyboard and continuous voice recognition (CVR)—for subjects classified by their decision style. In addition, the data for this experiment was compared to results of a similar experiment that used a discrete word recognition system. Dependent measures were task completion time, accuracy, keystroke count, correction count, and user confidence for spreadsheet tasks. Results, using a speaker-dependent continuous voice recognizer, showed that for both simple data input and more complex analytical problems, subjects did not perform more effectively using CVR compared to keyboard. In addition, a subject's decision style was found not to interact with CVR for an effect on performance. Compared to earlier discrete word recognition results, CVR tended to shorten the time to complete a spreadsheet analysis task.

Dysarthric speakers' intelligibility and speech characteristics in relation to computer speech recognition

This research examined the use of the DragonDictate speech recognition system as a writing aid by 10 speakers with spastic dysarthria due to cerebral palsy with the ultimate goal of establishing prescription guidelines. The objectives were to (1) determine how intelligibility relates to recognition success; (2) determine those characteristics of speech, fluency, and voice that affect the accuracy level of the voice recognizer; and (3) investigate individual profiles of recognition over a series of dictations among high- and low-intelligibility dysarthric speakers versus nondisabled controls. Correlations between intelligibility measures and recognition success measures were strong. The total number of voice features correlated moderately with the number of readings needed to reach 80% recognition levels and with variability in recognition. The frequency of pauses showed a strong positive association with the number of readings subjects needed to reach 80% recognition. The frequency of nonspeech sounds was...

Airborne Message Entry by Voice Recognition

This report presents the results of an experiment that measured performance in a simulated ASW message entry task with two modes of data input—vocal and manual. The subjects (Ss) were 12 Naval enlisted men. The independent variable was message data entry mode—vocal or manual. The dependent variables were: time to enter 20 lines of text, data entry errors that were corrected by the Ss, and errors that remained undetected. All Ss were trained to use the voice recognition system with a 100 word vocabulary set. The task was for the S to read one line of message text from a display and then re-enter the text below the displayed text via either voice recognizer or keyboard until 20 lines of text had been entered. Keyboard entry was found to be slightly faster (11%) than voice recognition input. While the number of initial errors (corrected) in the vocal input mode was over three times greater than the number for manual input, the remaining input errors (uncorrected) were about the same.