PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Speech Recognition Software”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5Linked to original sources
PubMed2005

Extending the evaluation of a computer system used as a microswitch for word utterances of persons with multiple disabilities.

BACKGROUND: Micro-switches can be vital tools to help individuals with extensive multiple disabilities acquire control of environmental stimulation. This study was aimed at extending the evaluation of a computer system used as a microswitch for word utterances with three participants with multiple disabilities. METHOD: Sets of 7 or 12 word utterances were used for the participants. The utterances were divided into three groups, which were exposed to intervention successively. During the intervention and a 2-month post-intervention check, the participants' emission of the target utterances led the system to present favourite, matching stimuli (i.e. provided that it recognized the utterances). RESULTS: Intervention data showed that (1) the participants increased the frequencies of the target utterances and (2) the computer system recognized approximately 80% of those utterances. These findings were maintained at the post-intervention check. An analysis of the levels of occurrence of individual utterances showed statistically significant differences among them, in line with the notions of preference and choice. CONCLUSIONS: The computer system was useful as a microswitch to enable access to favourite stimuli. There is a need to improve the accuracy of the system with respect to its recognition of the participants' utterances.

Adolescent↗
PubMed2005

Pilot of a novel cervical screening intervention: interactive voice response.

OBJECTIVE: This paper describes the pilot of an interactive voice response (IVR) cervical screening brief advice interface. METHOD: In September 2000, 5,000 households within Cessnock, New South Wales, were contacted by an IVR system. The system randomly selected eligible women aged 18-69, assessed their cervical screening status and provided a tailored message about cervical screening. A computer-assisted telephone interview (CATI) evaluated intervention acceptability. Participation rates, participant demographics, IVR use by women, and IVR costs were also examined. RESULTS: 549 (12%) women listened to the IVR call and 587 (59%) completed the CATI interview. IVR listeners under-represented younger and older women, and over-represented women who were married, and of Aboriginal descent. The majority of women (98%) found IVR easy to answer and 90% agreed IVR reminders were a good idea. Unscreened and older women used the system more than screened or younger women. CONCLUSIONS: The IVR call was successfully used by a large number of women to obtain information on cervical cancer screening, with higher use by recognised risk groups. IMPLICATIONS: These results suggest that an IVR reminder might be useful to increase cervical screening rates since those most at risk also used IVR the most. The potential for linking IVR to a database to issue reminders in the same manner as Pap Test Register postal reminders needs to be explored.

Adult↗
PubMed2005

Statistical properties of infant-directed versus adult-directed speech: insights from speech recognition.

Previous studies have shown that infant-directed speech ('motherese') exhibits overemphasized acoustic properties which may facilitate the acquisition of phonetic categories by infant learners. It has been suggested that the use of infant-directed data for training automatic speech recognition systems might also enhance the automatic learning and discrimination of phonetic categories. This study investigates the properties of infant-directed vs. adult-directed speech from the point of view of the statistical pattern recognition paradigm underlying automatic speech recognition. Isolated-word speech recognizers were trained on adult-directed vs. infant-directed data sets and were tested on both matched and mismatched data. Results show that recognizers trained on infant-directed speech did not always exhibit better recognition performance; however, their relative loss in performance on mismatched data was significantly less severe than that of recognizers trained on adult-directed speech and presented with infant-directed test data. An analysis of the statistical distributions of a subset of phonetic classes in both data sets showed that this pattern is caused by larger class overlaps in infant-directed speech. This finding has implications for both automatic speech recognition and theories of infant speech perception.

Adult↗
PubMed2004

Text-independent speaker verification using Minimal Resource Allocation Networks.

This paper presents a text-independent speaker verification system based on an online Radial Basis Function (RBF) network referred to as Minimal Resource Allocation Network (MRAN). MRAN is a sequential learning RBF, in which hidden neurons are added or removed as training progresses. LP-derived cepstral coefficients are used as feature vectors during training and verification phases. The performance of MRAN is compared with other well-known RBF and Elliptical Basis Function (EBF) based speaker verification methods in terms of error rates and computational complexity on a series of speaker verification experiments. The experiments use data from 258 speakers from the phonetically balancedcontinuous speech corpus TIMIT. The results show that MRAN produces comparable error rates to other methods with much less computational complexity.

Algorithms↗
PubMed2004

A neural network model with feature selection for Korean speech act classification.

A speech act is a linguistic action intended by a speaker. Speech act classification is an essential part of a dialogue understanding system because the speech act of an utterance is closely tied with the user's intention in the utterance. We propose a neural network model for Korean speech act classification. In addition, we propose a method that extracts morphological features from surface utterances and selects effective ones among the morphological features. Using the feature selection method, the proposed neural network can partially increase precision and decrease training time. In the experiment, the proposed neural network showed better results than other models using comparatively high-level linguistic features. Based on the experimental result, we believe that the proposed neural network model is suitable for real field applications because it is easy to expand the neural network model into other domains. Moreover, we found that neural networks can be useful in speech act classification if we can convert surface sentences into vectors with fixed dimensions by using an effective feature selection method.

Asian People↗
PubMed2005

Virtual vocalization stimuli for investigating neural representations of species-specific vocalizations.

Most studies investigating neural representations of species-specific vocalizations in non-human primates and other species have involved studying neural responses to vocalization tokens. One limitation of such approaches is the difficulty in determining which acoustical features of vocalizations evoke neural responses. Traditionally used filtering techniques are often inadequate in manipulating features of complex vocalizations. Furthermore, the use of vocalization tokens cannot fully account for intrinsic stochastic variations of vocalizations that are crucial in understanding the neural codes for categorizing and discriminating vocalizations differing along multiple feature dimensions. In this work, we have taken a rigorous and novel approach to the study of species-specific vocalization processing by creating parametric "virtual vocalization" models of major call types produced by the common marmoset (Callithrix jacchus). The main findings are as follows. 1) Acoustical parameters were measured from a database of the four major call types of the common marmoset. This database was obtained from eight different individuals, and for each individual, we typically obtained hundreds of samples of each major call type. 2) These feature measurements were employed to parameterize models defining representative virtual vocalizations of each call type for each of the eight animals as well as an overall species-representative virtual vocalization averaged across individuals for each call type. 3) Using the same feature-measurement that was applied to the vocalization samples, we measured acoustical features of the virtual vocalizations, including features not explicitly modeled and found the virtual vocalizations to be statistically representative of the callers and call types. 4) The accuracy of the virtual vocalizations was further confirmed by comparing neural responses to real and synthetic virtual vocalizations recorded from awake marmoset auditory cortex. We found a strong agreement between the responses to token vocalizations and their synthetic counterparts. 5) We demonstrated how these virtual vocalization stimuli could be employed to precisely and quantitatively define the notion of vocalization "selectivity" by using stimuli with parameter values both within and outside the naturally occurring ranges. We also showed the potential of the virtual vocalization stimuli in studying issues related to vocalization categorizations by morphing between different call types and individual callers.

Acoustic Stimulation↗
PubMedDate not supplied

Interactive voice response reminder effects on preventive service utilization.

This study evaluated the effects of interactive voice response (IVR) system reminders to managed care organization (MCO) members to obtain mammograms, Papanicolaou (Pap) tests, and influenza immunizations. The MCO identified 3 member cohorts and sent IVR reminders to get preventive services. Analyses employed claims data to examine relationships between IVR reminders and preventive service use 5 to 9 months post-intervention among members without prior utilization. Multivariate logistic regressions controlling for age, gender (for influenza immunizations), and risk stratum confirmed hypothesized relationships between intervention and preventive services: mammograms, odds ratio (OR) = 1.263 (95% confidence interval [CI] = 1.104, 1.444); Pap tests, OR = 1.241 (1.107, 1.391); influenza immunizations, OR = 2.072 (1.665, 2.580). IVR reminders are associated with higher rates of mammograms, Pap tests, and influenza immunizations. Study limitations include unknown generalizability of results and possible self-selection. There is justification for more IVR interventions and research to enhance MCO members' preventive service utilization.

Adult↗
PubMed2006

Usability of a barcode scanning system as a means of data entry on a PDA for self-report health outcome questionnaires: a pilot study in individuals over 60 years of age.

BACKGROUND: Throughout the medical and paramedical professions, self-report health status questionnaires are used to gather patient-reported outcome measures. The objective of this pilot study was to evaluate in individuals over 60 years of age the usability of a PDA-based barcode scanning system with a text-to-speech synthesizer to collect data electronically from self-report health outcome questionnaires. METHODS: Usability of the system was tested on a sample of 24 community-living older adults (7 men, 17 women) ranging in age from 63 to 93 years. After receiving a brief demonstration on the use of the barcode scanner, participants were randomly assigned to complete two sets of 16 questions using the bar code wand scanner for one set and a pen for the other. Usability was assessed using directed interviews with a usability questionnaire and performance-based metrics (task times, errors, sources of errors). RESULTS: Overall, participants found barcode scanning easy to learn, easy to use, and pleasant. Participants were marginally faster in completing the 16 survey questions when using pen entry (20/24 participants). The mean response time with the barcode scanner was 31 seconds longer than traditional pen entry for a subset of 16 questions (p = 0.001). The responsiveness of the scanning system, expressed as first scan success rate, was less than perfect, with approximately one-third of first scans requiring a rescan to successfully capture the data entry. The responsiveness of the system can be explained by a combination of factors such as the location of the scanning errors, the type of barcode used as an answer field in the paper version, and the optical characteristics of the barcode scanner. CONCLUSION: The results presented in this study offer insights regarding the feasibility, usability and effectiveness of using a barcode scanner with older adults as an electronic data entry method on a PDA. While participants in this study found their experience with the barcode scanning system enjoyable and learned to become proficient in its use, the responsiveness of the system constitutes a barrier to wide-scale use of such a system. Optimizing the graphical presentation of the information on paper should significantly increase the system's responsiveness.

Activities of Daily Living↗
PubMed2005

Speech-based interaction in multitask conditions: impact of prompt modality.

Speech-based interaction is often recognized as appropriate for hands-busy, eyes-busy multitask situations. The objective of this study was to explore prompt-guided speech-based interaction and the impact of prompt modality on overall performance in such situations. A dual-task paradigm was employed, with tracking as a primary task and speech-based data input as a secondary task. There were three tracking conditions: no tracking, basic, and difficult tracking. Two prompt modalities were used for the speech interaction: a dialogue with spoken prompts and a dialogue with visual prompts. Data entry duration was longer with the speech prompts than with the visual prompts, regardless of whether or not there was tracking or its level of difficulty. However, when tracking was difficult, data entry duration was similar for both spoken and visual prompts. Tracking performance was also affected by the prompt modality, with poorer performance obtained when the prompts were visual. The findings are discussed in terms of multiple resource theory and the possible implications for speech-based interactions in multitask situations. Actual or potential applications of this research include the design of speech-based dialogues for multitask situations such as driving and other hands-busy, eyes-busy situations.

Adolescent↗
PubMed2004

Address entry while driving: speech recognition versus a touch-screen keyboard.

A driving simulator experiment was conducted to determine the effects of entering addresses into a navigation system during driving. Participants drove on roads of varying visual demand while entering addresses. Three address entry methods were explored: word-based speech recognition, character-based speech recognition, and typing on a touch-screen keyboard. For each method, vehicle control and task measures, glance timing, and subjective ratings were examined. During driving, word-based speech recognition yielded the shortest total task time (15.3 s), followed by character-based speech recognition (41.0 s) and touch-screen keyboard (86.0 s). The standard deviation of lateral position when performing keyboard entry (0.21 m) was 60% higher than that for all other address entry methods (0.13 m). Degradation of vehicle control associated with address entry using a touch screen suggests that the use of speech recognition is favorable. Speech recognition systems with visual feedback, however, even with excellent accuracy, are not without performance consequences. Applications of this research include the design of in-vehicle navigation systems as well as other systems requiring significant driver input, such as E-mail, the Internet, and text messaging.

Accidents, Traffic↗
PubMed2004

Usage, performance, and satisfaction outcomes for experienced users of automatic speech recognition.

This paper presents a variety of outcomes data from 24 experienced users of automatic speech recognition (ASR) as a means of computer access. To assess usage and satisfaction, we conducted an in-person survey interview. For those participants who had a choice of computer input methods, 48% reported using ASR for 25% or less of their computer tasks, while 37% used ASR for more than half of their computer tasks. Users' overall satisfaction with ASR was somewhat above neutral (averaging 63 out of 100), and the most important role for ASR was as a means of reducing upper-limb pain and fatigue. To measure user performance, we asked users to perform a series of word processing and operating system tasks with their ASR systems. For 18 of these users, performance without speech was also measured. The time for nontext tasks was significantly slower with speech (p < 0.05). The average rate for entering text was no different with or without speech. Text entry rate with speech varied widely, from 3 to 32 words per minute, as did recognition accuracy, from 72% to 94%. Users who had the best performance tended to be those who employed the best correction strategies while using ASR.

Adolescent↗
PubMedDate not supplied

Effect of high-frequency spectral components in computer recognition of dysarthric speech based on a Mel-cepstral stochastic model.

Computer speech recognition of individuals with dysarthria, such as cerebral palsy patients, requires a robust technique that can handle conditions of very high variability and limited training data. In this study, a hidden Markov model (HMM) was constructed and conditions investigated that would provide improved performance for a dysarthric speech (isolated word) recognition system intended to act as an assistive/control tool. In particular, we investigated the effect of high-frequency spectral components on the recognition rate of the system to determine if they contributed useful additional information to the system. A small-size vocabulary spoken by three cerebral palsy subjects was chosen. Mel-frequency cepstral coefficients extracted with the use of 15 ms frames served as training input to an ergodic HMM setup. Subsequent results demonstrated that no significant useful information was available to the system for enhancing its ability to discriminate dysarthric speech above 5.5 kHz in the current set of dysarthric data. The level of variability in input dysarthric speech patterns limits the reliability of the system. However, its application as a rehabilitation/control tool to assist dysarthric motor-impaired individuals such as cerebral palsy subjects holds sufficient promise.

Artificial Intelligence↗
PubMed2006

Enabling persons with multiple disabilities to choose among environmental stimuli and request stimulus repetitions through microswitch and computer technology.

This study assessed microswitch and computer technology to enable two participants with multiple disabilities, 32 and 19 years of age, to choose among environmental stimuli and request their repetition whenever they so desired. Within each session, 18 stimuli (12 preferred and 6 nonpreferred) were available. For each stimulus, a computer system provided a sample of 3 sec. duration. During the intervention, participants' vocal responding in relation to a stimulus sample activated a microswitch and a computer system turned on that stimulus for 15-30 sec. When participants did not vocally respond, the computer system paused briefly and then presented the next scheduled stimulus sample. When participants vocally responded at the end of a stimulus presentation, the stimulus was repeated. Intervention data showed that the participants learned to choose preferred stimuli and bypass nonpreferred ones as well as to request repetitions of the preferred stimuli. This performance was maintained at a 1-mo. postintervention check and transferred to sets of stimuli not used in the intervention. The implications of these results were discussed.

Adult↗
PubMed2005

The delayed trigger voice key: an improved analogue voice key for psycholinguistic research.

Many researchers rely on analogue voice keys for psycholinguistic research. However, the triggering of traditional simple threshold voice keys (STVKs) is delayed after response onset, and the delay duration may vary depending on initial phoneme type. The delayed trigger voice key (DTVK), a stand-alone electronic device that incorporates an additional minimum signal duration parameter, is described and validated in two experiments. In Experiment 1, recorded responses from a nonword naming task were presented to the DTVK and an STVK. As compared with hand-coded reaction times from visual inspection of the waveform, the DTVK was more accurate than the STVK, overall and across initial phoneme type. Rastle and Davis (2002) showed that an STVK more accurately detected an initial [s] when it was followed by a vowel than when followed by a consonant. Participants' responses from that study were presented to the DTVK in Experiment 2, and accuracy was equivalent for initial [s] in vowel and consonant contexts. Details for the construction of the DTVK are provided.

Equipment Design↗
PubMed2006

The use of additionally trained sonographers as ultrasound practitioners: our first-year experience.

OBJECTIVE: Two sonographers were trained to help manage an abrupt, permanent increase in the number of ultrasound examinations in our department. Called "ultrasound practitioners," they functioned as physician assistants and triaged 20 to 30 cases per day, allowing the cases to be batch read at a formal reading at day's end. We report our first-year experience with this program. METHODS: Two sonographers with 10 and 30 years of experience, respectively, were trained to triage and dictate cases. Once trained, they triaged the cases of 20 to 30 patients per day. Reports were predictated with voice recognition technology. A radiologist was always readily available to provide support, and consultation with a radiologist was always obtained for the infrequent verbal reports that were requested. Reports from the practitioner were graded subjectively on a 4-point scale for the first year, according to the modification required at formal readout (A, no change; B, minor change not affecting patient care; C, moderate change not affecting care in a dramatic way; and D, major change markedly affecting care). RESULTS: Practitioner 1 monitored the examinations of 2858 patients. The graded report results were as follows: A, 96.2%; B, 3.5%; C, 0.3%; and D, 0.00%. Practitioner 2 monitored the examinations of 2825 patients. The graded report results were as follows: A, 96.1%; B, 3.6%; C, 0.2%; and D, 0.00%. There were no category D reports. CONCLUSIONS: The results far exceeded expectations, with a very low rate of category B and C reports and an absence of category D reports. The practitioners allowed the cases of 20 to 30 patients to be batch read by the existing radiologist staff at the end of the day.

Clinical Competence↗
PubMed2003

Electronic access to care system: improving patient's access to clinical information through an Interactive Voice Response (IVR) system and Web portal.

Our clinical providers spend an estimated four hours weekly answering phone messages from patients. Our nurses spend five to ten hours weekly on returning phone calls. Most of this time is spent conveying recent clinical results, reviewing with patients the discharge instructions such as consults or studies ordered during the office visits, and handling patients' requests for medication renewals. Over time this will lead to greater patients' dissatisfaction because of lengthy waiting time and lack of timely access to their medical information. This would also lead to greater nursing and providers' dissatisfaction because of unreasonable work load.

Access to Information↗