Speech Recognizers


 


Crowd translator: On building localized speech recognizers through micropayments [PDF] from mit.edu J Ledlie, B Odero, E Minkov, I Kiss… – ACM SIGOPS Operating …, 2010 – dl.acm.org Abstract We present a method to expand the number of languages covered by simple  speech recognizers. Enabling speech recognition in users’ primary languages greatly  extends the types of mobile-phone-based applications available to people in developing … Cited by 9 – Related articles – All 18 versions

[PDF] Practical evaluation of speech recognizers for virtual human dialogue systems [PDF] from psu.edu X Yao, P Bhutada, K Georgila, K Sagae… – Proceedings of …, 2010 – Citeseer Abstract We perform a large-scale evaluation of multiple off-the-shelf speech recognizers  across diverse domains for virtual human dialogue systems. Our evaluation is aimed at  speech recognition consumers and potential consumers with limited experience with … Cited by 3 – Related articles – View as HTML – All 7 versions

A multi-FPGA 10x-real-time high-speed search engine for a 5000-word vocabulary speech recognizer EC Lin… – Proceeding of the ACM/SIGDA international …, 2009 – dl.acm.org Abstract Today’s best quality speech recognition systems are implemented in software.  These systems fully occupy the resources of a high-end server to deliver results at real-time  speed: each hour of audio requires a significant fraction of an hour of computation for … Cited by 11 – Related articles

Speech Recognizer with Dynamic Alternative Path Search and Its Performance Evaluation T Morimoto… – Intelligent Automation and Computer …, 2010 – Springer For a middle-size (around 1,000 words) vocabulary speech recognition, a Finite State  Automaton (FSA) language model is widely used. However, defining a FSA model with  sufficient coverage and consistency requires much human effort. We already proposed a … Related articles – All 2 versions

[PDF] Transforming features to compensate speech recogniser models for noise [PDF] from pitt.edu RC Van Dalen, F Flego… – Proc. Interspeech, 2009 – cs.pitt.edu Abstract To make speech recognisers robust to noise, either the features or the models can  be compensated. Feature enhancement is often fast; model compensation is often more  accurate, because it predicts the corrupted speech distribution. It is therefore able, for … Cited by 10 – Related articles – All 7 versions

OpenMP-based parallel implementation of a continuous speech recognizer on a multi-core system [PDF] from mirlab.org K You, Y Lee… – Acoustics, Speech and Signal …, 2009 – ieeexplore.ieee.org Abstract We have implemented a 20,000-word continuous speech recognizer on a multi- core based system. A fine grain parallel processing approach is employed for good  scalability, and the OpenMP library is used for enhanced portability. In the emission … Cited by 15 – Related articles – All 5 versions

Improved topic classification and keyword discovery using an HMM-based speech recognizer trained without supervision MH Siu, H Gish, A Chan… – … Annual Conference of the …, 2010 – isca-speech.org In our previous publication, we presented a new approach to HMM training, viz., training  without supervision. We used an HMM trained without supervision for transcribing audio into  self-organized units (SOUs) for the purpose of topic classification. In this paper we report … Cited by 5 – Related articles

Why word error rate is not a good metric for speech recognizer training for the speech translation task? [PDF] from microsoft.com X He, L Deng… – Acoustics, Speech and Signal …, 2011 – ieeexplore.ieee.org Abstract Speech translation (ST) is an enabling technology for cross-lingual oral  communication. A ST system consists of two major components: an automatic speech  recognizer (ASR) and a machine translator (MT). Nowadays, most ASR systems are … Cited by 5 – Related articles – All 7 versions

Building a highly accurate Mandarin speech recognizer with language-independent technologies and language-dependent modules [PDF] from washington.edu MY Hwang, G Peng, M Ostendorf… – Audio, Speech, and …, 2009 – ieeexplore.ieee.org Abstract We describe a system for highly accurate large-vocabulary Mandarin speech  recognition. The prevailing hidden Markov model based technologies are essentially  language independent and constitute the backbone of our system. These include … Cited by 6 – Related articles – All 8 versions

A 1000-word vocabulary, speaker-independent, continuous live-mode speech recognizer implemented in a single FPGA [PDF] from cornell.edu EC Lin, K Yu, RA Rutenbar… – … of the 2007 ACM/SIGDA 15th …, 2007 – dl.acm.org Abstract The Carnegie Mellon In Silico Vox project seeks to move best-quality speech  recognition technology from its current software-only form into a range of efficient all- hardware implementations. The central thesis is that, like graphics chips, the application is … Cited by 36 – Related articles – All 11 versions

Speech recognizer optimization under speed constraints I Bulyko – Eleventh Annual Conference of the International …, 2010 – isca-speech.org We present an efficient algorithm for optimizing parameters of a speech recognizer aimed at  obtaining maximum accuracy at a specified decoding speed. This algorithm is not tied to any  particular decoding architecture or type of tunable parameter being used. It can also be … Cited by 2 – Related articles

A real-time FPGA-based 20 000-word speech recognizer with optimized DRAM access Y Choi, K You, J Choi… – Circuits and Systems I: …, 2010 – ieeexplore.ieee.org Abstract A real-time hardware-based large vocabulary speech recognizer requires high  memory bandwidth. We have developed a field-programmable-gate-array (FPGA)-based 20  000-word speech recognizer utilizing efficient dynamic random access memory (DRAM) … Cited by 7 – Related articles – All 3 versions

GMM-Based Matching Ability Measurement of a Speech Recognizer and a Feature Set HK Kim… – Future Communication, Computing, Control and …, 2012 – Springer In this work, we propose a Gaussian mixture model-based recognizer selection method to  overcome the acoustic mismatch between training and testing environments of a speech  recognition system. The method evaluates the preference of a system over other for a …

HEAR: an hybrid episodic-abstract speech recognizer [PDF] from pitt.edu S Demange… – Tenth Annual Conference of the …, 2009 – isca-speech.org This paper presents a new architecture for automatic continuous speech recognition called  HEAR-Hybrid Episodic-Abstract speech Recognizer. HEAR relies on both parametric  speech models (HMMs) and episodic memory. We propose an evaluation on the Wall … Cited by 7 – Related articles – All 14 versions

[PDF] Spoken Term Detection Using Multiple Speech Recognizers’ Outputs at NTCIR-9 SpokenDoc STD subtask [PDF] from nii.ac.jp H Nishizaki, H Furuya, S Natori… – Proceedings of the Ninth …, 2011 – research.nii.ac.jp ABSTRACT This paper describes spoken term detection (STD) with false detection control  using a phoneme transition network (PTN) derived from multiple speech recognizers’  outputs at NTCIR-9 SpokenDoc STD subtask. Using the output of multiple speech … Cited by 1 – View as HTML

Japanese Spoken Term Detection Using Syllable Transition Network Derived from Multiple Speech Recognizers’ Outputs S Natori, H Nishizaki… – Eleventh Annual Conference …, 2010 – isca-speech.org This paper proposes a spoken term detection using syllable transition network (STN)  derived from multiple speech recognizers. An STN is similar to a sub-word based confusion  network, which is derived from the output of a speech recognizer. The one we proposed is … Cited by 4 – Related articles

[CITATION] Large Vocabulary Continuous Recognition/Search-Building A Highly Accurate Mandarin Speech Recognizer With Language-Independent Technologies … MY Hwang, G Peng, M Ostendorf, W Wang… – IEEE transactions on audio, …, 2009

Incorporating acoustical modelling of phone transitions in an hybrid ANN/HMM speech recognizer [PDF] from inesc-id.pt A Abad, J Neto – Ninth Annual Conference of the International …, 2008 – isca-speech.org Speech recognition based on connectionist approaches is one of the most successful  alternatives to widespread Gaussian systems. One of the main claims against hybrid  recognizers is the increased complexity for context-dependent phone modelling, which is … Cited by 16 – Related articles – All 3 versions

A Commercial Car Navigation System using Korean Large Vocabulary Automatic Speech Recognizer [PDF] from hokudai.ac.jp SJ Lee, H Chung, JG Park… – … APSIPA ASC 2009 …, 2009 – eprints2008.lib.hokudai.ac.jp In this paper, a Korean large vocabulary speech recognizer for an embedded car navigation  device is introduced. The proposed speech recognizer identifies 450k point-of-interests  within a resource-limited device without serious performance degradation under severe … Cited by 4 – Related articles – All 5 versions

Context dependent modelling approaches for hybrid speech recognizers [PDF] from inesc-id.pt A Abad, T Pellegrini, I Trancoso… – … Annual Conference of the …, 2010 – isca-speech.org Speech recognition based on connectionist approaches is one of the most successful  alternatives to widespread Gaussian systems. One of the main claims against hybrid  recognizers is the increased complexity for context-dependent phone modeling, which is a … Cited by 4 – Related articles – All 3 versions

Unsupervised training of an HMM-based Speech Recognizer for Topic Classification [PDF] from pitt.edu H Gish, M Siu, A Chan… – Tenth Annual Conference of …, 2009 – isca-speech.org HMM-based Speech-To-Text (STT) systems are widely deployed not only for dictation tasks  but also as the first processing stage of many automatic speech applications such as spoken  topic classification. However, the necessity of transcribed data for training the HMMs … Cited by 1 – Related articles – All 3 versions

Optimization of Dereverberation Parameters based on Likelihood of Speech Recognizer [PDF] from pitt.edu R Gomez… – Tenth Annual Conference of the …, 2009 – isca-speech.org Speech recognition under reverberant condition is a difficult task. Most dereverberation  techniques used to address this problem enhance the reverberant waveform independent  from that of the speech recognizer. In this paper, we improve the conventional Spectral … Cited by 7 – Related articles – All 7 versions

Improvement of a speech recognizer for standardized medical assessment of children’s speech by integration of prior knowledge T Bocklet, A Maier, U Eysholdt… – … Workshop (SLT), 2010 …, 2010 – ieeexplore.ieee.org Abstract Speech recognition of children is a more difficult task than speech recognition of  adults. This problem is amplified for children with articulation disorders like cleft lip and  palate (CLP). In this work we improved our automatic speech recognition system by … Related articles

[PDF] Adaptation of a speech recognizer to singing voice [PDF] from psu.edu A Mesaros… – Proceedings of 17th European Signal Processing …, 2009 – Citeseer ABSTRACT This paper studies the speaker adaptation techniques that can be applied for  adapting a speech recognizer to singing voice. Maximum likelihood linear regression  (MLLR) techniques are studied, with specific details in choosing the number and types of … Cited by 5 – Related articles – View as HTML – All 9 versions

An Effective Feature Compensation Scheme Tightly Matched with Speech Recognizer Employing SVM-Based GMM Generation W Kim, JW Suh… – Eleventh Annual Conference of the …, 2010 – isca-speech.org This paper proposes an effective feature compensation scheme to address a real-life  situation where clean speech database is not available for Gaussian Mixture Model (GMM)  training for a model-based feature compensation method. The proposed scheme employs … Related articles

Adapting the Acoustic Model of a Speech Recognizer for Varied Proficiency Non-Native Spontaneous Speech Using Read Speech with Language-Specific … [PDF] from pitt.edu K Zechner, D Higgins, R Lawless… – … Annual Conference of …, 2009 – isca-speech.org This paper presents a novel approach to acoustic model adaptation of a recognizer for non- native spontaneous speech in the context of recognizing candidates’ responses in a test of  spoken English. Instead of collecting and then transcribing spontaneous speech data, a … Cited by 1 – Related articles – All 5 versions

Building a highly accurate Mandarin speech recognizer [PDF] from columbia.edu MY Hwang, G Peng, W Wang… – … , 2007. ASRU. IEEE …, 2007 – ieeexplore.ieee.org Abstract We describe a highly accurate large-vocabulary continuous Mandarin speech  recognizer, a collaborative effort among four research organizations. Particularly, we build  two acoustic models (AMs) with significant differences but similar accuracy for the … Cited by 22 – Related articles – All 21 versions

Transforming Features to Compensate Speech Recogniser Models for Noise RC Dalen, F Flego… – Tenth Annual Conference of the …, 2009 – isca-speech.org To make speech recognisers robust to noise, either the features or the models can be  compensated. Feature enhancement is often fast; model compensation is often more  accurate, because it predicts the corrupted speech distribution. It is therefore able, for … Related articles

A model distance maximizing framework for speech recognizer-based speech enhancement B BabaAli, H Sameti… – AEU-International Journal of Electronics …, 2011 – Elsevier This paper has presented a novel discriminative parameter calibration approach based on  the model distance maximizing (MDM) framework to improve the performance of our  previously-proposed method based on spectral subtraction (SS) in a likelihood- … Related articles

Scalable architecture of tone classification function for tonal speech recognizer J Chaiwongsai, W Chiracharit… – Intelligent Signal …, 2010 – ieeexplore.ieee.org Abstract Tone classification function is used for improving recognition accuracy in tonal  speech recognizer (TONE-SPEC). Although average magnitude difference function (AMDF)  is generally used to find pitch period of fundamental frequency, there are many frame- … Related articles

A Robust Bengali Continuous Speech Recognizer Using Triphone and Trigram Language Model S Mandal, B Das, P Mitra… – Contemporary Computing, 2011 – Springer In this paper we introduce a robust Bengali Automatic Speech Recognition (ASR) system  which covers most of the commonly spoken words. This ASR system converts standard  Bengali continuous speech to Bengali Unicode with a decent accuracy rate. The existing … Related articles

Can non-linear readout nodes enhance the performance of reservoir-based speech recognizers? F Triefenbach… – Informatics and Computational …, 2011 – biblio.ugent.be Ga onmiddellijk naar paginanavigatie. Can non-linear readout nodes enhance the performance of reservoir-based speech recognizers? … Title, Can non-linear readout nodes enhance the performance of reservoir-based speech recognizers? Publication Status, inpress. … Cached – All 2 versions

Estimation of Two-to-One Forced Selection Intelligibility Scores by Speech Recognizers Using Noise-Adapted Models K Kondo… – Eleventh Annual Conference of the …, 2010 – isca-speech.org We attempted to estimate subjective scores of the Japanese Diagnostic Rhyme Test (DRT),  a two-to-one forced selection speech intelligibility test, using automatic speech recognizers  with language models that force one of the words in the word-pair. The acoustic models … Cited by 2 – Related articles

[PDF] Maximum Likelihood Training and Adaptation of Embedded Speech Recognizers for Mobile Environments [PDF] from etri.re.kr Y Cho… – ETRI journal, 2010 – etrij.etri.re.kr For the acoustic modek of embedded speech recognition systems, hidden Markov models  (HMMs) are usually quantized and the original full space distributions are represented by  combinations of a few quantized distribution prototypes. We propose a maximum … Related articles – All 4 versions

Performance improvement in multiple-model speech recognizer under noisy environments JH Yoon… – Structural, Syntactic, and Statistical Pattern …, 2010 – Springer Multiple-model speech recognizer has been shown to be quite successful in noisy speech  recognition. However, its performance has usually been tested using the general speech  front-ends which do not incorporate any noise adaptive algorithms. For the accurate … Related articles – All 2 versions

Open vocabulary spoken document retrieval by subword sequence obtained from speech recognizer G Kuriki, Y Itoh, K Kojima… – … , 2008. SLT 2008. …, 2008 – ieeexplore.ieee.org Abstract We present a method for open vocabulary retrieval based on a spoken document  retrieval (SDR) system using subword models. The present paper proposes a new approach  to open vocabulary SDR system using subword models which do not require subword … Related articles – All 2 versions

The benefit obtained from visually displayed text from an automatic speech recognizer during listening to speech presented in noise AA Zekveld, SE Kramer, JM Kessens… – Ear and …, 2008 – journals.lww.com Objectives: The aim of this study was to evaluate the benefit that listeners obtain from  visually presented output from an automatic speech recognition (ASR) system during  listening to speech in noise. Design: Auditory-alone and audiovisual speech reception … Cited by 6 – Related articles – All 5 versions

[PDF] Combination of 3 Types of Speech Recognizers for Anaphora Resolution [PDF] from kyutech.ac.jp K Shimada, N Tanamachi… – Proceedings of the 24nd …, 2010 – pluto.ai.kyutech.ac.jp Abstract. In this paper, we propose a method for anaphora resolution in speech  understanding for a livelihood support robot. For robust speech recognition, we combine two  types of speech recognizers; a large vocabulary continuous speech recognizer (LVCSR) … Cited by 1 – Related articles – View as HTML

Development of a Speech Recognizer with the Tecnovoz Database [PDF] from pp.ua J Lopes, C Neves, A Veiga, A Maciel… – … Processing of the …, 2008 – Springer This paper describes the development of a robust speech recognition using a database  collected in the scope of the Tecnovoz project. The speech recognition system is speaker  independent, robust to noise and operates in a small footprint embedded hardware … Cited by 8 – Related articles – All 6 versions

Speed improvements in a Missing Data-based speech recogniser by Gaussian selection Y Wang… – 2009 – Citeseer Abstract Speech recognition performance in noisy environments such as cars is degraded  due to the mismatch between the feature vector and the speech model. To improve the noise  robustness, we apply Missing Data Techniques (MDT) to Hidden Markov Models (HMM) … Cached – All 2 versions

Design and implementation of a Bayesian network speech recognizer P Wiggers, L Rothkrantz… – Text, Speech and Dialogue, 2011 – Springer In this paper we describe a speech recognition system implemented with generalized  dynamic Bayesian networks (dbn s). We discuss the design of the system and the features of  the underlying toolkit we constructed that makes efficient processing of speech and … Cited by 3 – Related articles – All 3 versions

Improving automatic speech recognizer of voice search using system combination T Li, W Xu, J Pan… – Fuzzy Systems and Knowledge …, 2009 – ieeexplore.ieee.org Abstract Voice search is the technology that enables users to access information using  spoken queries. Automatic speech recognizer (ASR) is one of the key modules for voice  search systems. However, the high error rate of the state-of-the-art large vocabulary … Cited by 1 – Related articles – All 4 versions

[PDF] An effective speech understanding method with a multiple speech recognizer based on output selection using edit distance [PDF] from kyutech.ac.jp K Shimada, S Horiguchi… – Proceedings of the 22nd …, 2008 – pluto.ai.kyutech.ac.jp Abstract. In this paper, we propose a simple and effective method for speech understanding.  The method incorporates some speech recognizers. We use two recognizers, a large  vocabulary continuous speech recognizer and a domain-specific speech recognizer. The … Cited by 5 – Related articles – View as HTML

An open-source speech recognizer for brazilian portuguese with a windows programming interface [PDF] from googlecode.com P Silva, P Batista, N Neto… – Computational Processing of the …, 2010 – Springer This work is part of the effort to develop a speech recognition system for Brazilian  Portuguese. The resources for the training and test stages of this system, such as corpora,  pronunciation dictionary, language and acoustic models, are publicly available. Here, an … Cited by 1 – Related articles – All 4 versions

The HMM Synthesis Algorithm of an Embedded Unified Speech Recognizer and Synthesizer [PDF] from pitt.edu G Strecha, M Wolff, F Duckhorn… – … Annual Conference of …, 2009 – isca-speech.org In this paper we present an embedded unified speech recognizer and synthesizer using  identical, speaker independent Hidden-Markov-Models. The system was prototypically  realized on a signal processor extended by a field programmable gate array. In a first … Related articles – All 3 versions

[PDF] FPGA-based implementation of a real-time 5000-word continuous speech recognizer [PDF] from eurasip.org Y Choi, K You… – Proc. 16th Eur. Signal Process. Conf, 2008 – eurasip.org ABSTRACT We have developed a hidden Markov model based 5000-word speaker  independent continuous speech recognizer using a Field-Programmable Gate Array  (FPGA). The feature extraction is conducted in software on a soft-core based CPU, while … Cited by 6 – Related articles – View as HTML – All 2 versions

[CITATION] Parametric Representations in HMM-based Speech Recognizers S Kanokphara – 2009 – University College Dublin Library Search

[PDF] Comparing Speech Recognizers Derived from Mono-and Multilingual Grammars [PDF] from univ-paris13.fr ME Santaholma – 2009 – www-lipn.univ-paris13.fr Abstract This paper examines the performance of multilingual parameterized grammar rules  on speech recognition. We present a performance comparison of two different types of  Japanese and English grammar-based speech recognizers. One system is derived from … Related articles – View as HTML – All 3 versions

[CITATION] Morpheme Conversion for Connecting Speech Recognizer and Language Analyzers in Unsegmented Languages K Imamura, T Izumi, K Sadamitsu, K Saito… – epos, 2011 Related articles

Combined static and dynamic variance adaptation for efficient interconnection of speech enhancement pre-processor with speech recognizer M Delcroix, T Nakatani… – Acoustics, Speech and …, 2008 – ieeexplore.ieee.org Abstract It is well known that automatic speech recognition performs poorly in presence of  noise or reverberation. Much research has been undertaken on model adaptation and  speech enhancement to increase the robustness of speech recognizers. Model adaptation … Cited by 3 – Related articles – All 3 versions

Applying a Multi-Voice Speech Recognizer to the BMIST Task [PDF] from dtic.mil GJ Gadbois – 2008 – DTIC Document Abstract: Speech recognition topics are explored. An orally driven user interface to form  filling was developed. Along the way, unsupervised adaptation methods, noise injection and  novel enrollment/model search methods were studied. On the strength of the discovered … Library Search – All 3 versions

Recognition of voice commands using adaptation of foreign language speech recognizer via selection of phonetic transcriptions R Maskeliunas… – Central European Journal of Engineering, 2011 – Springer Abstract In recent years various commercial speech recognizers have become available.  These recognizers provide the possibility to develop applications incorporating various  speech recognition techniques easily and quickly. All of these commercial recognizers are … Related articles – All 4 versions

N-gram language models in JLASER neural network speech recognizer M Konopi´k, I Habernal… – Applied Electronics (AE), …, 2010 – ieeexplore.ieee.org Abstract In our recent research we have discovered that neural networks can be more  efficient in speech recognition than the state of the art approach based on Gaussian  mixtures. This statement is valid only for small corpora, however, many applications do not … Related articles

Rapid adaptation using linear spectral transformation for embedded speech recognisers Y Cho… – Electronics Letters, 2008 – ieeexplore.ieee.org Abstract Embedded speech recognisers are typically used in unknown mobile environments  where the acoustic conditions frequently change. Since a large amount of adaptation data is  not usually available for such environments, the adaptation methods for the acoustic … Cited by 1 – Related articles – All 3 versions

Applications of Virtual-Evidence based Speech Recognizer Training [PDF] from washington.edu A Subramanya… – Ninth Annual Conference of the …, 2008 – isca-speech.org We present two applications of our previously proposed virtualevidence (VE) based speech  recognizer training algorithm [1, 2]. The first relates to two-pass training where  segmentations obtained during the first pass are used as VE to train the subsequent pass. … Cited by 2 – Related articles – All 7 versions

[PDF] Building a visual speech recognizer [PDF] from tudelft.nl KF Driel – Delft University of Technology, 2009 – repository.tudelft.nl Summary This thesis describes how an automatic lip reader was realized. Visual speech  recognition is a precondition for more robust speech recognition in general. The  development of the software comprised the following steps: gathering of training data, … Cited by 1 – Related articles – View as HTML – All 3 versions

Appropriate Farsi speech recognizer for commanding robots:(Performance evaluation of correlation-based and model-based classifiers for a Farsi isolated word … A Rashedi… – Signal Processing (ICSP), 2010 …, 2010 – ieeexplore.ieee.org Abstract In this research, two different classifier categories, correlation-based and neural  network-based, are investigated for a Farsi isolated word recognizer commanding robotic  system. Correlation-based category is divided to time and frequency domains. Moreover, …

[CITATION] Speech Recognizer Adaptation A Maier – 2008 – VDM Verlag Dr. Müller, Saarbrücken … Cited by 2 – Related articles

Uncertainty in training large vocabulary speech recognizers [PS] from washington.edu A Subramanya, C Bartels, J Bilmes… – … Speech Recognition & …, 2007 – ieeexplore.ieee.org Abstract We propose a technique for annotating data used to train a speech recognizer. The  proposed scheme is based on labeling only a single frame for every word in the training set.  We make use of the virtual evidence (VE) framework within a graphical model to take … Cited by 7 – Related articles – All 6 versions

Improve detection performance of speech recognizer in an automotive environment A Sapru, R Lakkundi… – Signals, Systems and …, 2008 – ieeexplore.ieee.org Abstract In-car speech recognition is a challenging area of research. The area has been  studied, and more of is concentrated on addressing the issues in acoustic echo cancellation  (AEC). In application such as a voice controlled car audio system, voice commands by the … Related articles

Virtual evidence for training speech recognizers using partially labeled data [PDF] from upenn.edu A Subramanya… – … Technologies 2007: The Conference of the …, 2007 – dl.acm.org Abstract Collecting supervised training data for automatic speech recognition (ASR) systems  is both time consuming and expensive. In this paper we use the notion of virtual evidence in  a graphical-model based system to reduce the amount of supervisory training data … Cited by 4 – Related articles – All 20 versions

cROVER: Context-augmented Speech Recognizer based on Multi-Decoders’ Output [PDF] from uwaterloo.ca MK Abida – 2011 – uwspace.uwaterloo.ca Abstract: The growing need for designing and implementing reliable voice-based human- machine interfaces has inspired intensive research work in the field of voice-enabled  systems, and greater robustness and reliability are being sought for those systems. … Related articles – All 5 versions

[PDF] An on-line speaker adaptation method for HMM-based speech recognizers [PDF] from u-szeged.hu A Bnhalmi… – Acta Cybernetica, 2008 – inf.u-szeged.hu Abstract In the past few years numerous techniques have been proposed to improve the  efficiency of basic adaptation methods like MLLR and MAP. These adaptation methods have  a common aim, which is to increase the likelihood of the phoneme models for a particular … Cited by 1 – Related articles – View as HTML – All 3 versions

Is a Speech Recognizer Useful for Characteristic Analysis of Classroom Lecture Speech? K Kobayashi, M Somiya, H Nishizaki… – … Annual Conference of …, 2008 – isca-speech.org This paper investigates whether a speech recognizer is useful for characteristic analysis of  lecture speech or not. It is important to pay attention to how we speak in order to effectively  convey the meaning of what we are trying to say to someone. We have examined how … Cited by 1 – Related articles

Multiplatform server-based speech recogniser generator for embedded systems R Krejci… – Applied Electronics (AE), 2011 …, 2011 – ieeexplore.ieee.org Abstract Automatic speech recognisers still more come in the practice of regular users of  computers and various devices. However, so far there is no speech recogniser on the  market in the form of an electronic component or a module that could be easily usable in …

[CITATION] ct. a1. Building a highly accurate mandarin speech recognizer MY Hwang – Proc. IEEE Automatic Speech Recognition and …, 2007 Cited by 2 – Related articles

Why is this Wrong?–Diagnosing Erroneous Speech Recognizer Output with a Two Phase Parser [PDF] from ffzg.hr B Ludwig… – Proceeding of the 2008 conference on ECAI 2008 …, 2008 – dl.acm.org Abstract A major problem of understanding language in spoken dialog systems is to detect  recognition errors in the output of a speech recognizer. Such a capability is the basis of  implementing repair strategies that allow a dialog system to handle communication about … Related articles – All 15 versions

A hierarchical multiple recognizer for robust speech understanding T Yokoyama, K Shimada… – PRICAI 2010: Trends in Artificial …, 2010 – Springer … The method incorporates some speech rec- ognizers. We use two types of recognizers; a large vocabulary continuous speech recognizer and a domain-specific speech recognizer. … Keywords: Multiple speech recognizer, Output selection, Hierarchical method. 1 Introduction … Cited by 1 – Related articles – All 2 versions

Feature Selection Algorithms for Creation of Multistream Speech Recognizers Y Kubo, S Okawa, A Kurematsu… – 2008 – isca-speech.org … Background • Multistream speech recognizers • Our objectives • Proposed method … Effective use of many features (high-dimensional features) is important • Multistream speech recognizers (MSRs) can utilize many features derived by multiple feature analysis methods … Related articles – All 2 versions

[CITATION] JLASER: An Automatic Speech Recognizer Written in Java T Pavelka… – Proc. of XII International Conference Speech and …, 2007 Cited by 4 – Related articles

[PDF] LPFAV2: a multi-modal database for developing continuous speech recognisers in assistive technology applications [PDF] from psu.edu A Moura, V Pêra… – 2008 – Citeseer 1School of Technology and Management, Polytechnic Institute of Bragança Quinta de Sta  Apolónia, Apartado 134, 5301 – 857 Bragança, Portugal phone: +351 273 303130, fax: 273  313051, email: moura@ipb.pt, web: www.estig.ipb.pt/ … 2LSS, DEEC, Faculty of … Related articles – View as HTML – All 6 versions

On chip realization of HMM speaker-independent isolated word speech recognizer W WAN… – Information Technology, 2008 – en.cnki.com.cn An embedded speaker-independent isolated word speech recognition system is designed  and realized in the Xilinx Virtex-II Pro platform. With the help of a modified real time voice  activity detection algorithm (VAD) based on the log. Energy acceleration associated with … Cited by 1 – Related articles – Cached – All 2 versions

Robust Romanian language automatic speech recognizer based on multistyle training [PDF] from wseas.us DP Munteanu… – WSEAS Transactions on Computer Research, 2008 – dl.acm.org Abstract This paper presents solutions for increasing environmental robustness of a  Romanian language continuous speech recognizer, previously developed. All state-of-the- art automatic speech recognizers (ASR) are data-driven and rely heavily on huge speech … Related articles – All 2 versions

A generic interface methodology for bridging application systems and speech recognizers SJ Peng… – Information, Communications & Signal …, 2007 – ieeexplore.ieee.org Abstract Application systems that utilize recognition technologies, such as speech  recognition, provide human-machine interface that could aid people more easily in  operating system device or help those who are physically unable to interact with … Cited by 1 – Related articles

Evaluating unsupervised data in isolated speech recognizer N Seman, SS Salleh… – … Engineering, 2008. ICCCE …, 2008 – ieeexplore.ieee.org Abstract This paper presents initial studies of applying isolated speech recognizer (ISR) on  different datasets of adultpsilas speech ISR is normally created for a targeted user or  language. Even though the targeted user is defined, there are often speakers who are … Related articles

A Speech Recognizer based on Multiclass SVMs with HMM-Guided Segmentation M Iglesias – 2010 – recolecta.net Resumen Automatic Speech Recognition (ASR) is essentially a problem of pattern  classification, however, the time dimension of the speech signal has prevented to pose ASR  as a simple static classification problem. Support Vector Machine (SVM) classifiers could … Cached

[CITATION] Optimization of a Speech Recognizer for Medical Studies on Children in Preschool and Primary School Age T Bocklet – 2007 – Diplomarbeit, Chair of Pattern … Cited by 2 – Related articles

Comparison of automatic speech recognizer SPHINX 3.6 and SPHINX 4.0 for creating systems in Slovak language J Vojtko, J Korosi… – Systems, Signals and Image …, 2008 – ieeexplore.ieee.org Abstract In this paper we discuss a topic of an automatic speech recognition system based  on a system SPHINX in various versions and configurations. We compare Sphinx version 3  and 4 for recognition of Slovak speech. Other comparison is focused on the type of a … Related articles

[CITATION] On chip realization of speaker-independent isolated word speech recognizer CY Zhang, CL Sun – Jisuanji …, 2007 – … Technology Institute, No. 26, PO Box …

Handling Phonetic Context and Speaker Variation in a Structure-Based Speech Recognizer [PDF] from microsoft.com D Yu, L Deng… – Eighth Annual Conference of the …, 2007 – isca-speech.org Recently we have developed a novel type of structure-based speech recognizer, which uses  parameterized, non-recursive” hidden” trajectory model of vocal tract resonances (VTR) or  formants to capture the dynamic structure of long-range speech coarticulation and … Related articles – All 6 versions

[CITATION] A syllable based continuous speech recognizer for indian languages A Lakshmi – 2007 – Master’s thesis, Indian Institute of … Cited by 2 – Related articles

[CITATION] Using a small development data set to build a robust dialectal Chinese speech recognizer LQ Liu, F Zheng… – Proc of INTERSPEECH, 2007 Cited by 2 – Related articles

[PDF] Robust Romanian language automatic speech recognizer [PDF] from wseas.us D Munteanu… – Proc. of The 6th WSEAS International …, 2007 – wseas.us Abstract:-In this paper there are presenting solutions for increasing environmental  robustness of a Romanian language continuous speech recognizer, previously built [1],[2]  as a man-machine dialogue system. Multistyle training strategy is used to train the … Cited by 1 – Related articles – View as HTML – All 2 versions

Speech recognition system having multiple speech recognizers N Endo – Acoustical Society of America Journal, 2007 – adsabs.harvard.edu Title: Speech recognition system having multiple speech recognizers. Authors: Endo, Norikazu. Publication: The Journal of the Acoustical Society of America, vol. 122, issue 5, p. 2515. Publication Date: 00/2007. Origin: CROSSREF. DOI: 10.1121/1.2801848. … All 2 versions

[PDF] Telephone-Based Spoken Dialog System Using HTK-based Speech Recognizer and VoiceXML [PDF] from ovgu.de KT Mengistu… – FORTSCHRITTE DER AKUSTIK, 2007 – iesk.ovgu.de Given the ubiquity of speech and the availability of the telephone coupled with the current  state of speech technology, it is realistic to claim unconditional access to information.  Telephone-based spoken dialog systems can take us closer to this claim of accessing … Cited by 1 – Related articles – View as HTML – BL Direct – All 7 versions

Using a Small Development Set to Build a Robust Dialectal Chinese Speech Recognizer [PDF] from tsinghua.edu.cn L Liu, TF Zheng, M Akabane… – … Annual Conference of …, 2007 – isca-speech.org To make full use of a small development data set to build a robust dialectal Chinese speech  recognizer from a standard Chinese speech recognizer (based on Chinese Initial/Final, IF),  a novel, simple but effective acoustic modeling method, named state-dependent phoneme … Related articles – All 5 versions

Response time reduction of speech recognizers using single gaussians [PDF] from kaist.ac.kr S Jeong, H Kim… – IEICE TRANSACTIONS ON …, 2007 – dspace.kaist.ac.kr Title Response Time Reduction of Speech Recinizers Using Single Gaussians Authors  JEONG, Sangbae; KIM, Hoirin; HAHN, Minsoo Keywords speech recognition; fast likelihood  computation Issue Date May-2007 Publisher Institute of Electronics, Information and … Related articles – BL Direct – All 15 versions

[PDF] Speech understanding in a multiple recognizer with an anaphora resolution process [PDF] from kyutech.ac.jp K Shimada, A Uzumaki… – Proceedings of the 11th …, 2009 – pluto.ai.kyutech.ac.jp … Utsuro et al. (2004) have obtained high accuracy by using some speech recognizers’ out- puts. … If we handle these different speech recognizers selectively and integratively, we re- alize a flexible and robust speech understanding method. … Cited by 4 – Related articles – View as HTML

Telecommunication system, speech recognizer, and terminal, and method for adjusting capacity for vocal commanding HU Roeck – Acoustical Society of America Journal, 2007 – adsabs.harvard.edu Title: Telecommunication system, speech recognizer, and terminal, and method for adjusting capacity for vocal commanding. Authors: Roeck, Hans-Ueli. Publication: The Journal of the Acoustical Society of America, vol. 121, issue 6, p. 3267. Publication Date: 00/2007. … All 2 versions

[PDF] Augmentation of Noise Free Speech Recognizer using Adaptive Microphone Array [PDF] from iete.org M MUKHOPADHYAY, A KUNDU… – The Institution of …, 2007 – iete.org Reverberation can be loosely defined as the effect an environment, eg a room, has on the  propagation of an acoustic signal produced within it. An acoustic signal will travel in a  straight line directly to a receiver located in the room. This is called the direct path. The … Related articles – View as HTML – All 4 versions

A novel emotion recognizer from speech using both prosodic and linguistic features M Suzuki, S Tsuchiya… – Knowledge-Based and Intelligent …, 2011 – Springer … Then, subtitles given by the speech recognizer are input for another emotion recognizer based on the “Association Mecha- nism.” It outputs a possible emotion by using only linguistic information. … Speech recognizers, such as the one employed in this system, are not perfect. … Related articles – All 2 versions

… , folyamatos beszédfelismero rendszer megvalósítási megoldásainak kutatása= Research on the construction of continuous speech recognizer for a Hungarian middle … [PDF] from mtak.hu K Vicsi, G Gordos, M Naszódi… – OTKA Kutatási Jelentések| …, 2007 – real.mtak.hu Abstract A 3 év alatt a tervnek megfeleloen az alábbi feladatokat végeztük el: 1.  Létrehoztunk egy általános, olvasott szövegu, magyar nyelvu beszédadatbázist, amely  irodai környezetben használható beszédfelismerok akusztikai-fonetikai modelljeinek … Related articles

Leveraging multimodal redundancy for dynamic learning, with SHACER, a speech and handwriting recognizer EC Kaiser – 2007 – en.scientificcommons.org … To combine information SHACER normalizes handwriting and speech recognizer out-puts by applying letter-to-sound and sound-to-letter transformations. SHACER then uses an articulatory-feature based distance metric to align handwriting to redundant speech. … Cited by 1 – Related articles – Cached – Library Search – All 2 versions

[PDF] PERANCANGAN PROGRAM APLIKASI SPEECH RECOGNIZER DENGAN SPECTROGRAM [PDF] from binus.ac.id AY STEFAN – 2008 – eprints.binus.ac.id Sebelumnya secara khusus penulis mengucap syukur dan terima kasih kepada Tuhan  Yang Maha Esa yang menyertai penulis dalam perjuangan mengatasi rintangan untuk  dapat menyelesaikan skripsi ini. Kuasa-Nya menjadi nyata saat penulis melihat …

Open vocabulary spoken document retrieval by subword sequence obtained from a speech recognizer ????, ????, ????… – … IEEE Workshop on …, 2008 – dspace.itri.aist.go.jp We present a method for open vocabulary retrieval based on a spoken document retrieval  (SDR) system using subword models. The present paper proposes a new approach to open  vocabulary SDR system using subword models which do not require subword recognition. … Cached – All 3 versions

Lithuanian speech recognition using the English recognizer [PDF] from mii.lt P Kasparaitis – Informatica, 2008 – IOS Press … Received: November 2007; accepted: February 2008 Abstract. The present work is concerned with speech recognition using a small or medium size vo- cabulary. The possibility to use the English speech recognizer for the recognition of Lithuanian was investigated. … Cited by 9 – Related articles – All 4 versions