Voice Input Research Articles

Testing the Use of Voice Input in a Smartphone Web Survey

We implemented an experiment within a smartphone web survey to explore the feasibility of using voice input (VI) options. Based on device used, participants were randomly assigned to a treatment or control group. Respondents in the iPhone operating system (iOS) treatment group were asked to use the dictation button, in which the voice was translated automatically into text by the device. Respondents with Android devices were asked to use a VI button which recorded the voice and transmitted the audio file. Both control groups were asked to answer open-ended questions using standard text entry. We found that the use of VI still presents a number of challenges for respondents. Voice recording (Android) led to substantially higher nonresponse, whereas dictation (iOS) led to slightly higher nonresponse, relative to text input. However, completion time was significantly reduced using VI. Among those who provided an answer, when dictation was used, we found fewer valid answers and less information provided, whereas for voice recording, longer and more elaborated answers were obtained. Voice recording (Android) led to significantly lower survey evaluations, but not dictation (iOS).

Interaction Strategies for Effective Augmented Reality Geo-Visualization: Insights from Spatial Cognition

Standalone augmented reality (AR) systems have great potential for interactive three-dimensional (3D) geo-visualization. Emerging head-worn AR technologies can display rich graphical imagery of large-scale environments and permit intuitive interaction through gestural and voice inputs. Yet, how users engage in geo-visualization through these interfaces and what interaction strategies yield the best spatial memory and navigation performance remain open questions. In the present empirical investigation, we related user interactions in a 3D geo-visualization application implemented on the Microsoft HoloLens AR system to virtual navigation outcomes. Informed by spatial cognitive theory, we uncovered interaction strategies during goal-oriented study of a 3D urban environment that predicted different aspects of navigation performance. Users who positioned the 3D city model to gain an overhead (i.e. survey perspective) view early on performed best when later following the route from memory. In contrast, consistent interaction switching (i.e. perspective transformation) during study predicted navigational efficiency when participants were unexpectedly tasked to return to the origin of the route. Individual differences also related to aspects of user interaction. Implications for 3D geo-visualization and navigation-assistive AR application design and suggestions for future directions are discussed.

A Novel User Interface for Text Dependent Human Voice Recognition System

In an effort to provide a more efficient representation of the speech signal, the application of the wavelet analysis is considered. This research presents an effective and robust method for extracting features for speech processing. Here, we proposed a novel user interface for Text Dependent Human Voice Recognition (TD-HVR) system. The proposed HVR model utilizes decimated bi-orthogonal wavelet transform (DBT) approach to extract the low level features from the given input voice signal, then the noise elimination will be done by band pass filtering followed by normalization for better quality of a voice signal and finally the formants of a train and test voices will be calculated by using the Additive Prognostication (AP) algorithm. Simulation results have been compared with the existing HVR schemes, and shown that the proposed user interface system has performed superior to the conventional HVR systems with an accuracy rate of approximately 99 %.

A Novel User Interface for Text Dependent Human Voice Recognition System

In an effort to provide a more efficient representation of the speech signal, the application of the wavelet analysis is considered. This research presents an effective and robust method for extracting features for speech processing. Here, we proposed a novel user interface for Text Dependent Human Voice Recognition (TD-HVR) system. The proposed HVR model utilizes decimated bi-orthogonal wavelet transform (DBT) approach to extract the low level features from the given input voice signal, then the noise elimination will be done by band pass filtering followed by normalization for better quality of a voice signal and finally the formants of a train and test voices will be calculated by using the Additive Prognostication (AP) algorithm. Simulation results have been compared with the existing HVR schemes, and shown that the proposed user interface system has performed superior to the conventional HVR systems with an accuracy rate of approximately 99 %.

The Key Technology of Speech Interaction Based on Deep Learning

Language recognition is essentially a pattern recognition problem, and all kinds of pattern recognition methods are used for language recognition research. For language recognition studied in this project, the highly effective speech feature parameters, acoustic model and language recognition classifier, the Chinese, Tibetan, Mongolian, uygur and training of acoustic model are established, so as to identify the voice input languages. In recent years, the deep learning theory has been applied in the field of speech recognition, and the DNN-HMM model has become the mainstream of acoustic modeling. This project studies Tibetan, Mongolian, Uygur emotion modeling of acoustic model and language model, training methods, studies the decoding algorithm, for continuous speech recognition of input speech, converts it to ethnic Chinese words.

The Use Of Voice in Opac Using Google API Voice Recognition & Speech Synthesis and Fullproof Algorithm as Faster Searching Device

The development of technology provides convenience to all human beings including public librarians. OPAC becomes a solution for library users to search books in a short time with accurate data in which they previously used a manual system. However, there are some users, especially users with certain disabilities, such as users who cannot see, are unable to use OPAC. The idea came up to make OPAC with voice as a substitute for another alternative of text input use. OPAC with voice has been used in several big libraries in Indonesia. One of them is in State Islamic University of Sunan Kalijogo. However, because there are still some shortcomings in terms of voice recognition in which one of them is the limited number of recognizable languages, so in this study, it will be discussed the use of Google API for voice recognition and voice synthesis which is then integrated with OPAC, and the algorithm used in the search is Fullproof algorithm. The methods used in this research were literature study, system design, prototype application creation, and application testing. From the test results, the application is able to search the data book in OPAC using voice input from various languages with a relatively fast searching time. Applications are also able to voice the book found in the form of sound.

A study of the facilitation of cross‐cultural understanding and intercultural sensitivity using speech‐enabled language translation technology

AbstractWe proposed speech‐enabled language translation (SELT) technology that consists of speech‐to‐text recognition (STR) and computer‐aided‐translation (CAT) systems to support intercultural communication. When the participants spoke in their native languages, the STR system generated texts from their voice input, and the CAT system translated the STR‐generated texts in foreign languages into English. We aimed to investigate whether our approach would be useful to facilitate cross‐cultural understanding and the intercultural sensitivity of the participants. Our results demonstrated that cross‐cultural learning literally took place during the learning activity and that the intercultural sensitivity of the participants improved. Furthermore, low intelligibility and accuracy rates were obtained for Mongolian, Filipino, Hindi, Mandarin, and Vietnamese, while high intelligibility and accuracy rates were obtained for Spanish, Russian and French during the SELT process. Finally, our results showed that the students highly valued the application of SELT as support for the cross‐cultural learning activity due to its ease of use, its ability to enable access to authentic content, and the collaboration that occurred among the participants. Based on our results, we provide some implications and useful suggestions to educators and researchers.

Electronic Health Record Interactions through Voice: A Review.

Usability problems in the electronic health record (EHR) lead to workflow inefficiencies when navigating charts and entering or retrieving data using standard keyboard and mouse interfaces. Voice input technology has been used to overcome some of the challenges associated with conventional interfaces and continues to evolve as a promising way to interact with the EHR. This article reviews the literature and evidence on voice input technology used to facilitate work in the EHR. It also reviews the benefits and challenges of implementation and use of voice technologies, and discusses emerging opportunities with voice assistant technology. We performed a systematic review of the literature to identify articles that discuss the use of voice technology to facilitate health care work. We searched MEDLINE and the Google search engine to identify relevant articles. We evaluated articles that discussed the strengths and limitations of voice technology to facilitate health care work. Consumer articles from leading technology publications addressing emerging use of voice assistants were reviewed to ascertain functionalities in existing consumer applications. Using a MEDLINE search, we identified 683 articles that were reviewed for inclusion eligibility. The references of included articles were also reviewed. Sixty-one papers that discussed the use of voice tools in health care were included, of which 32 detailed the use of voice technologies in production environments. Articles were organized into three domains: Voice for (1) documentation, (2) commands, and (3) interactive response and navigation for patients. Of 31 articles that discussed usability attributes of consumer voice assistant technology, 12 were included in the review. We highlight the successes and challenges of voice input technologies in health care and discuss opportunities to incorporate emerging voice assistant technologies used in the consumer domain.

Smart wheelchair using android smartphone for physically disabled people

This paper describes the development of a smart wheelchair system with voice recognition and touch controlled using an embedded system. An android application is developed and installed on the android smartphone. The system is divided into two main modes: voice recognition mode and touch mode. For the voice recognition mode, elderlies or physically disabled people (users) can provide the voice input, for example, “go”, “reverse”, “turn to the left”, “turn to the right” and “stop”. The wheelchair will move according to the command given. For the touch mode, the user can select the specified direction displayed within the four quadrants on the screen of the android smartphone to control the wheelchair. An Arduino Uno is used to execute all commands. The MD30C motor driver and HC05 Bluetooth module are used in this system. This system is designed to save time and energy of the user.

The recognition of female voice based on voice registers in singing techniques in real-time using hankel transform method and macdonald function

A singer doesn’t just recite the lyrics of a song, but also with the use of particular sound techniques to make it more beautiful. In the singing technique, more female have a diverse sound registers than male. There are so many registers of the human voice, but the voice registers used while singing, among others, Chest Voice, Head Voice, Falsetto, and Vocal fry. Research of speech recognition based on the female’s voice registers in singing technique is built using Borland Delphi 7.0. Speech recognition process performed by the input recorded voice samples and also in real time. Voice input will result in weight energy values based on calculations using Hankel Transformation method and Macdonald Functions. The results showed that the accuracy of the system depends on the accuracy of sound engineering that trained and tested, and obtained an average percentage of the successful introduction of the voice registers record reached 48.75 percent, while the average percentage of the successful introduction of the voice registers in real time to reach 57 percent.

An evaluation of multimodal interaction techniques for 3D layout constraint solver in a desktop-based virtual environment

We propose a new approach to the 3D layout problems based on the integration of constraint programming and virtual reality interaction techniques. Our method uses an open-source constraint solver integrated in a popular 3D game engine. We designed multimodal interaction techniques for the system, based on gesture and voice input. We conducted a user study with an interactive task of laying out room furniture to compare and evaluate the mono- and multimodal interaction techniques. Results showed that voice command provided the best performance and was most preferred by participants, based on the analysis of both objective and subjective data. Results also revealed that there was no significant difference between the voice and multimodal input (voice and gesture). Our original approach opens the way to multidisciplinary theoretical work and promotes the development of high-level applications for the VR applications.

Giving Respondents Voice? The Feasibility of Voice Input for Mobile Web Surveys

More and more respondents are completing Web surveys on mobile devices. Some studies have found that responses to open narrative questions are affected by use of mobile devices, especially smartphones. Given the rise of voice input on smartphones in general, it is worth investigating the feasibility of voice input for answering open questions on smartphones. The goal of this research note is to study (1) the extent to which people already use voice input on their mobile devices (for “everyday” life), (2) the extent to which they would be willing to use voice input to answer open questions in Web surveys, and (3) the factors that may affect the willingness to use voice input. Based on data from a survey in Spain in 2016, we found that around half of our sample already uses voice input (at least sometimes) and that also around half of the sample is willing to use it to answer open narrative questions if this was possible. Several variables were found to have a significant association with the willingness to use voice input: the use of voice input in everyday life, the frequency of Internet access through a smartphone, and the perceived difficulty of answering the current survey.

RETRACTED ARTICLE: Research and Experiment of Intelligent Natural Language Processing Algorithms

Natural language processing is mainly divided into two parts: speech processing and word processing. The level of word processing is mainly studied. Natural language processing is divided into lexical analysis, syntax analysis and semantic analysis. Aiming at the scope of the language ambiguity and thesaurus in the field of smart home, the maximal matching algorithm is used to segment the natural language. Then, through the way of template matching, semantic comprehension finally forms the code form that can control the home node. In the system applied in this paper, the speech is processed into words through the existing voice input function of the mobile terminal. Then, the control instruction is obtained through the language processing method. The processed data is communicated to the server via socket. The server sends the data to the home node through the Zigbee protocol. Finally, control of home appliances is achieved.

SHOW

Smart watch is becoming a new gateway through which people stay connected and track everyday activities, and text-entry on it is becoming a frequent need. With the two de facto solutions: tap-on-screen and voice input, text-entry on the watch remains a tedious task because 1. Tap-on-screen is error prone due to the small screen; 2. Voice input is strongly constrained by the surroundings and suffers from privacy leak. In this paper, we propose SHOW, which enables the user to input as they handwrite on horizontal surfaces, and the only requirement is to use the elbow as the support point. SHOW captures the gyroscope and accelerometer traces and deduces the user's handwriting thereafter. SHOW differs from previous work of gesture recognition in that: 1. it employs a novel rotation injection technique to substantially reduce the effort of data collection; 2. it does not require whole-arm posture, hence is better suited to space-limited places (e.g. vehicles). Our experiments show that SHOW can effectively generate 60 traces from one real handwriting trace and achieve high accuracy at 99.9% when recognizing the 62 different characters written by 10 volunteers. Furthermore, having more screen space after removing the virtual keyboard, SHOW can display 4x candidate words for autocompletion. Aided by the tolerance of character ambiguity and accurate character recognition, SHOW achieves over 70% lower mis-recognition-rate, 43% lower no-response-rate in both daily and general purposed text-entry scenarios, and 33.3% higher word suggestion coverage than the tap-on-screen method using a virtual QWERTY keyboard.

Smart Home for the Abled and Disabled Aged Persons in Africa: Design and Development

Smart home is an intelligent home network whose services and infrastructure are based information and communication technology. It is an intelligent use of wireless technology for the implementation of automated home electrical/electronic appliances and devices, domestic tasks, and security devices, for the wellbeing of the occupant. However, the development of smart homes has focused mainly on building for the young and able-bodied persons for the purpose of enhancing home comfort, energy saving, and security, without considering the aged and disabled persons in Africa. The Smart Home System were implemented in existing home environments, with the automation focusing on recognition of voice commands and uses low-power RF wireless communication modules along with a microcontroller for message passing. The system consists of voice recognition module for voice input and recognition. The control signal of the recognized command is transmitted to the microcontroller, and the microcontroller uses this signal to control home appliances. The designed system was tested and the result showed that the level of system response is higher in dead home than in a live home, hence, the system performance is affected by the environmental noise and the accents of un-trained aged persons are not supported by it. This system is most suitable for the aged and the disabled persons living alone and is intended to control all lights and electrical appliances inside a house using voice commands. In this work a home automation control system based on voice recognition was designed, implemented and experimental tests done and results were obtained.

Speech controlled home mechanization framework using android gadgets

This task displays the development Speech Controlled Home Mechanization Framework Using Android Gadgets of home apparatuses in light-weight of voice order utilizing humanoid. This technique has been meant to help and provides the help to senior and unfit individual’s reception. Recent voice based applications provide acknowledges the voice contribution from the advanced transportable. During this venture, the voice input has been caught by the ports and might be sent to the Arduino Uno. HC 05 module in Arduino Uno got the flag and handled the information flag to manage the two power sockets and fan. The proposed framework expected to manage electrical devices with general user friendly interface and easy transnational. In this project we have a tendency to gift associate humanoid OS based mostly application for smartphone that speaks with the fan through mobile phone persistently to manage the FAN speed. The humanoid stage assumes a key half to holds a most extreme range of users once contrasted with all different stage. We have got exhibited up to twenty meter of vary to manage the house apparatuses by suggests that of Bluetooth.

RANCANG BANGUN ROBOT HUMANOID PENARI GENDING SRIWIJAYA MENGGUNAKAN MODUL EASYVR3

The purpose of this final project to produce humanoid robot that can perform Gending Sriwijaya dance movement by using speech recognition module. There is an asynchronization between the music accompaniment detection movement of humanoid robot on Gending Sriwijaya dance at Kontes Robot Seni Tari Indonesia 2017. Use of speech recognition moduls input system given the solution of the problem.System requirements were divided into hardware and software included robot acuator components suh as 25 servo motors, robot controllers such as CM-530 and Arduino Mega, EasyVR3 speech recognition module, battery. There three software used to complied humanoid robot. EasyVR Commander software for the detection and synchronization of voice input, Roboplus software for making movements on humanoid robot, Arduino IDE software for programming of humanoid robot movement control system. The design of this humanoid robot system is divided into 3 parts, input of the system is Gending Sriwijaya music which is detected by the speech recognition module, the controlling process of the system is Arduino Mega and CM-530, and the output of the system is the movement of humanoid robot which is formed from the combination of rotation Angle servo motor. The result of this final project is successfully working humanoid robot that can perform Gending Sriwijaya dance movement by using speech recognition module EasyVR3. Furthermore, has successfully worked the program of detection and synchronization of voice, program control system movement of humanoid robot in dance Gending Sriwijaya in the form of movement walkready, sembah buka, cubung bawah, cubung atas, silang, transisi berdiri ke duduk (sembah duduk), borobudur, transisi duduk ke berdiri (sembah diri), ukel benang , sembah tutup, dan jalan. Keywords: humanoid robot, speech recognition, Gending Sriwijaya dance

Applications of speech-to-text recognition and computer-aided translation for facilitating cross-cultural learning through a learning activity: issues and their solutions

In this study, 21 university students, who represented thirteen nationalities, participated in an online cross-cultural learning activity. The participants were engaged in interactions and exchanges carried out on Facebook® and Skype® platforms, and their multilingual communications were supported by speech-to-text recognition (STR) and computer-aided translation (CAT) systems. The participants spoke in their native languages, and the STR system generated texts from their voice input. The CAT system then simultaneously translated the STR-texts into English. The aim of this study was to examine the accuracy rates of STR and CAT processes for different languages during intercultural communication. We also explored issues associated with these processes and how they were addressed in the study context. In addition, an attempt was made to determine whether or not our learning activity as supported by STR and CAT technologies facilitated cross-cultural learning. Our results showed that the lowest STR accuracy rate was for Belizean English whereas the highest STR accuracy rate was for French and Hindi. The lowest CAT accuracy rate was for Mongolian and Filipino, and the highest was for Spanish, Russian, and French. Seven issues associated with the STR process and ten issues associated with the CAT process were identified. The participants employed ten workarounds to address the STR-related issues and thirteen workarounds to address the CAT-related issues. We refer to a workaround as a method used by the participants to overcome a limitation related to either STR or CAT. Finally, our results demonstrated that cross-cultural learning took place; the participants understood and could explain foreign traditions to others and could also compare foreign traditions with their own. Based on our results, we made several suggestions and provided implications for the teaching and research community.

A voice command detection system for aerospace applications

Nowadays, according to ever-increasing volumes of audio content, audio processing is a vital need. In the aerospace field, voice commands could be used instead of data commands in order to speed up the command transmission, help crewmembers to complete their tasks by allowing hands-free control of supplemental equipment and as a redundant system for increasing the reliability of command transmission. In this paper, a voice command detection (VCD) framework is proposed for aerospace applications, which decodes the voice commands to comprehensible and executable commands, in an acceptable speed with a low false alarm rate. The framework is mainly based on a keyword spotting method, which extracts some pre-defined target keywords from the input voice commands. The mentioned keywords are input arguments to the proposed rule-based language model (LM). The rule-based LM decodes the voice commands based on the input keywords and their locations. Two keyword spotters are trained and used in the VCD system. The phone-based keyword spotter is trained on TIMIT database. Then, speaker adaptation methods are exploited to modify the parameters of the trained models using non-native speaker utterances. The word-based keyword spotter is trained on a database prepared and specialized for aerospace applications. The experimental results show that the word-based VCD system decodes the voice commands with true detection rate equal to 88% and false alarm rate equal to 12%, in average. Additionally, using speaker adaptation methods in the phone-based VCD system improves the true detection and false alarm rates about 21% and 21%, respectively.

A Focus Group Study of Blind Voters in Alachua County

Of the 15.6 million people with disabilities who voted in the 2012 United States Presidential Election, one- third had difficulties voting at the precinct. This statistic is one of the reasons work continues on Prime III, an all-accessible electronic voting system, designed to be used by all voters, regardless of ability or disability. To describe the current state of voting for disabled voters and identify issues Prime III can further address, focus groups were conducted with the Alachua County Council of the Blind in Florida. Four focus groups were run with 18 participants in total. Preliminary results find that while the current accessible voting machines satisfactorily allow disabled voters to cast their votes, there are features and functions Prime III can provide that are not available now such as voice input, universal access, and privacy. Next steps include conducting focus groups with disabled students at the University of Florida.