Papers by Anne-marie Öster
A computer-aided analytical speech perception test and a helpful training program have been devel... more A computer-aided analytical speech perception test and a helpful training program have been developed. The test is analytical and seeks to evaluate the ability to perceive a range of sound contrasts used in the Swedish language. The test is tailored for measurements with chil- dren, who have not yet learnt to read, by using easy speech stimuli, words selected on
In this paper we present recent results on the development of the SynFace lip synchronized talkin... more In this paper we present recent results on the development of the SynFace lip synchronized talking head towards multi-linguality, varying signal conditions and noise robustness in the Hearing at Home project. We then describe the large scale hearing impaired user studies carried out for three languages. The user tests focus on measuring the gain in Speech Reception Threshold in Noise when using SynFace, and on measuring the effort scaling when using SynFace by hearing impaired people. Preliminary analysis of the results does not show significant gain in SRT or in effort scaling. But looking at inter-subject variability, it is clear that many subjects benefit from SynFace especially with speech with stereo babble noise.
In this paper we present recent results on the development of the SynFace lip synchronized talkin... more In this paper we present recent results on the development of the SynFace lip synchronized talking head towards multi-linguality, varying signal conditions and noise robustness in the Hearing at Home project. We then describe the large scale hearing impaired user studies carried ...
International Journal of Speech Technology, 2000
The development of an audiovisual pronunciation teaching and training method and software system ... more The development of an audiovisual pronunciation teaching and training method and software system is discussed in this article. The method is designed to help children with speech and hearing disorders gain better control over their speech production. The teaching method is drawn up for progression from individual sound preparation to practice of sounds in sentences for four languages: English, Swedish, Slovenian, and Hungarian. The system is a general language-independent measuring tool and database editor. This database editor makes it possible to construct modules for all participant languages and for different sound groups. Two modules are under development for the system in all languages: one for teaching and training vowels to hearing-impaired children and the other for correction of misarticulated fricative sounds. In the article we present the measuring methods, the used distance score calculations of the visualized speech spectra, and problems in the evaluation of the new multimedia tool.
Πρακτικά Διεθνούς Συνεδρίου «Μάθηση και Διδασκαλία της Ελληνικής ως μητρικής και ως δεύτερης Γλώσσας», 2001
This paper relates to the contribution of computer technology in the teaching of pronunciation an... more This paper relates to the contribution of computer technology in the teaching of pronunciation and speech production in general. This technology is being used to help in the teaching of the pronunciation of a language, as well as in the production of various aspects of speech by language impaired people. Interactive software enables speakers to acquire an intuitive understanding of the relation between speech sounds and the articulatory behaviour that produces them, while the spectral representation of speech sounds provides visual feedback which is instrumental in improving speech production. A lot of phonetic research focuses on the aforementioned areas.
SPECO is an on-going European project (INCO-COPERNICUS,1998-2001) which is developing an audio-visual pronunciation teaching and training system for hearing and speech handicapped children. It is based on a product-oriented approach and incorporates an auditory model imitating the human peripheral auditory system during speech perception. Time-warping algorithms have been used to account for duration differences. The limits within which pronunciation can be characterised as normal are set according to measurements of isolated phonemes produced by at least 30 children and rated for acceptability by expert listeners.
SPECO is currently being developed for Hungarian, English, Slovenian and Swedish, and can be adapted to any European language through a language-independent database editor and measuring system. The two main modules of the project involve vowel and consonant support progressing from loudness and pitch exercises to syllable, word and phrase practice. Preliminary experiments for the evaluation of SPECO have shown great improvement in deaf children’s speech.
It is hoped that Greek will be one of the languages that SPECO will include in the future. The basic idea of this project can also be utilised to build a pronunciation teaching tool for learners of Greek as a second or foreign language.
In this paper, we present results on diode lasers in the wavelength range between 715 nm and 840 ... more In this paper, we present results on diode lasers in the wavelength range between 715 nm and 840 nm with Al-free QWs which are embedded in a high-quality AlGaAs LOC broadened waveguide structure with low optical loss and a small vertical far field divergence. The laser structures were grown by LP- MOVPE. We studied tensile-strained GaAsP-QWs as well as compressively
2000 IEEE International Symposium on Compound Semiconductors. Proceedings of the IEEE Twenty-Seventh International Symposium on Compound Semiconductors (Cat. No.00TH8498), 2000
This paper reports on the optimization of red vertical-cavity surface-emitting laser (VCSEL) diod... more This paper reports on the optimization of red vertical-cavity surface-emitting laser (VCSEL) diodes grown by metalorganic vapour phase epitaxy. The VCSEL structure has an GaInP/AlGaInP multiquantum well active zone sandwiched between AlGaAs/AlAs distributed Bragg reflectors (DBRs). We present results on the optimization of the electrical resistance of the p:DBR and the wavelength tuning. Using these optimized parameters pulsed laser operation
This paper presents new techniques used to produce visual maps for training Swedish fricatives fo... more This paper presents new techniques used to produce visual maps for training Swedish fricatives for hearing-impaired clients, and presents results from testing one such map with the speech of three hearing-impaired subjects. The results demonstrate the utility of the mapping techniques for visually portraying correct and deviant pronunciations and indicate the potential for use in actual clinical practice. The study
This study has been performed in order to test the man- machine interface of a computer-based spe... more This study has been performed in order to test the man- machine interface of a computer-based speech training aid named ARTUR with the main feature that it can give suggestions on how to improve articulation. Two user groups were involved: three children aged 9-14 with extensive experience of speech training, and three children aged 6. All children had general language
Proceedings of the 7th international ACM SIGACCESS conference on Computers and accessibility - Assets '05, 2005
This study has been performed in order to test the human-machine interface of a computer-based sp... more This study has been performed in order to test the human-machine interface of a computer-based speech training aid named ARTUR with the main feature that it can give suggestions on how to improve articulation. Two user groups were involved: three children aged 9-14 with extensive experience of speech training, and three children aged 6. All children had general language disorders.
CHI '06 extended abstracts on Human factors in computing systems - CHI EA '06, 2006
... Anne-Marie Öster Centre for Speech Technology KTH SE-100 44 Stockholm, Sweden annemarie@speec... more ... Anne-Marie Öster Centre for Speech Technology KTH SE-100 44 Stockholm, Sweden annemarie@speech.kth.se Hedvig Kjellström Interaction and Presentation Laboratory Computer-assisted pronunciation training, virtual tutor, feedback, Wizard of Oz, usability. ...
Conference Proceedings. LEOS '97. 10th Annual Meeting IEEE Lasers and Electro-Optics Society 1997 Annual Meeting, 1996
Abstract Summary form only given. In summary we have developed a diode laser with a low threshold... more Abstract Summary form only given. In summary we have developed a diode laser with a low threshold current density and a high efficiency using a compressively strained Al-free InGaAsP QW embedded in AlGaAs waveguides. Using the well established MOVPE for ...
European Conference on eHealth, 2007
In this papert he concepta nd architecture of am onitoring anda ssistant system forpeoplewithhear... more In this papert he concepta nd architecture of am onitoring anda ssistant system forpeoplewithhearing deficiencies will be describeda nd afirst userstudy will be presented. Core of this concept is theimplementation of algorithmsthata re usuallyu sed in hearinga idsa tacentral Home Informationa nd Communication (HIC)p latform.T hisp latform integrates formerly separated devices likeT V, telephone,a nd homea utomation. Thereby, theT Vb ecomest he centralh uman machinei nterfaceo ft he householdl inking together most of thea coustici nputsl istenedt ob yt he hearing impaired in his/her homee nvironment. Usingt he central HICp latformt he acoustic inputsc an be amplifieda nd adapteda ccordingt ot he specific hearingi mpairmenta nd thei ndividualt hresholds of hearing. These thresholds will be estimatedd uringt he initiald iagnosis at thea udiologist and transmittedtothe HICplatform. SupportiveAudioSignalProcessing (SASP) algorithms areused to fine-tune theamplification according to thea cousticsituationof thel ivingr oom.T hesef ine-tuning, frequent testso fs peechu nderstanding and monitoringofthe individual adjustment of volumeand sound controllerare used as afeedbackt ot he audiologist.I nt hisw ay thet ransition fromm ildt om oderateo r severehearinglosscan be supervised,the useofhearing aidscan be motivated and theadjustmentofhearing aids canbeimprovedand adaptedtothe individualsituationofthe patient.The ideasofthissound processinga nd alertingsystemathome, thesocalled Hearinga tHomeSystem( HaH) were presentedt o62hearinga id usersa nd non-users( "young-old" to "old-old"; mean age: 69yrs.)o ft hreec ountries (Netherlands,Sweden, andGermany) with adifferentdegreeofhearing loss (from slight to moderate). The first resultso ft he user requirement study showed,t hat most of theu sers will benefit from theH aH environments ystem, andt he results arev ery useful forf urther research andd evelopment, also in ther ealm of telemonitoringapplications.
2010 IEEE Radio Frequency Integrated Circuits Symposium, 2010
We report a compact 802.11b/g/n MIMO SoC with fully integrated transceiver, on-chip PMU including... more We report a compact 802.11b/g/n MIMO SoC with fully integrated transceiver, on-chip PMU including dc-dc converters, PHY, MAC, PCIe and a non-volatile memory. The transceiver includes on-chip PA, LNA and T/R switch. Fabricated in 90nm standard digital CMOS technology, this IC consumes 663/878mW (Rx/Tx 54Mbps) with an area of approx 33mm2. A peak saturated power of 24dBm is achieved at
The development of an audio-visual pronunciation teaching and training method and software system... more The development of an audio-visual pronunciation teaching and training method and software system is discussed in this paper. The method is designed to help children with speech and hearing disorders gain better control over their speech production. The teaching method is drawn up for progression from individual sound preparation to practice of sounds in sentences, for four languages: English, Swedish, Slovenian and Hungarian.
Fifth International Conference on Spoken …, 1998
This contribution presents part of the work initiated at the CTT for the development of speech te... more This contribution presents part of the work initiated at the CTT for the development of speech technology to assist non-native speakers learn Swedish. This s tudy focuses mainly on the automatic location o f mispronunciations at a phonetic level. We first describe the database we created for this work and then report on the reliability of several phonetic scores to
Uploads
Papers by Anne-marie Öster
SPECO is an on-going European project (INCO-COPERNICUS,1998-2001) which is developing an audio-visual pronunciation teaching and training system for hearing and speech handicapped children. It is based on a product-oriented approach and incorporates an auditory model imitating the human peripheral auditory system during speech perception. Time-warping algorithms have been used to account for duration differences. The limits within which pronunciation can be characterised as normal are set according to measurements of isolated phonemes produced by at least 30 children and rated for acceptability by expert listeners.
SPECO is currently being developed for Hungarian, English, Slovenian and Swedish, and can be adapted to any European language through a language-independent database editor and measuring system. The two main modules of the project involve vowel and consonant support progressing from loudness and pitch exercises to syllable, word and phrase practice. Preliminary experiments for the evaluation of SPECO have shown great improvement in deaf children’s speech.
It is hoped that Greek will be one of the languages that SPECO will include in the future. The basic idea of this project can also be utilised to build a pronunciation teaching tool for learners of Greek as a second or foreign language.
SPECO is an on-going European project (INCO-COPERNICUS,1998-2001) which is developing an audio-visual pronunciation teaching and training system for hearing and speech handicapped children. It is based on a product-oriented approach and incorporates an auditory model imitating the human peripheral auditory system during speech perception. Time-warping algorithms have been used to account for duration differences. The limits within which pronunciation can be characterised as normal are set according to measurements of isolated phonemes produced by at least 30 children and rated for acceptability by expert listeners.
SPECO is currently being developed for Hungarian, English, Slovenian and Swedish, and can be adapted to any European language through a language-independent database editor and measuring system. The two main modules of the project involve vowel and consonant support progressing from loudness and pitch exercises to syllable, word and phrase practice. Preliminary experiments for the evaluation of SPECO have shown great improvement in deaf children’s speech.
It is hoped that Greek will be one of the languages that SPECO will include in the future. The basic idea of this project can also be utilised to build a pronunciation teaching tool for learners of Greek as a second or foreign language.