
Robert EklundLinköping University | LiU · Department of Culture and Communication (IKK)
Robert Eklund
Associate Professor (Professor) in Language, Culture and Phonetics and Associate Professor (Docent/Habilitation) in Computational Linguistics; MA in Musicology
About
122
Publications
39,170
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
680
Citations
Introduction
Additional affiliations
August 2005 - January 2006
International Computer Science Institute (ICSI)
Position
- PostDoc Position
Description
- Postdoc programme on an AMI grant. On leave from TeliaSonera; Automatic Meeting Analysis
Publications
Publications (122)
Objective
A comprehensive understanding of how vocal tract dimensions vary among different types of loud voice productions has not yet been fully formed. This study aims to expand the existing knowledge on the topic.
Methods
Three trained professional singers together practiced the vocal techniques underlying Opera and Kulning singing styles for o...
This study investigates domestic cat meows in different contexts and mental states. Measures of fundamental frequency (f0) and duration as well as f0 contours of 780 meows from 40 cats were analysed. We found significant effects of recording context and of mental state on f0 and duration. Moreover, positive (e.g. affiliative) contexts and mental st...
In this paper we study segment prolongations (PRs), a type of disfluency sometimes included under the term "hesitation disfluencies", in Hebrew. PRs have previously been studied in a number of other languages within a comprehensive speech disfluency framework, which is applied to Hebrew in the current study. For the purpose of this study we defined...
In this paper we study segment prolongations (PRs), a type of disfluency sometimes included under the term "hesitation disfluencies", in Hebrew. PRs have previously been studied in a number of other languages within a comprehensive speech disfluency framework, which is applied to Hebrew in the current study. For the purpose of this study we defined...
This study investigates domestic cat meows in different contexts and mental states. Measures of fundamental frequency (f0) and duration as well as f0 contours of 780 meows from 40 cats were analysed. We found significant effects of recording context and of mental state on f0 and duration. Moreover, positive (e.g. affiliative) contexts and mental st...
This study investigates domestic cat meows in different contexts and mental states. Measures of fundamental frequency (f0) and duration as well as f0 contours of 780 meows from 40 cats were analysed. We found significant effects of recording context and of mental state on f0 and duration. Moreover, positive (e.g. affiliative) contexts and mental st...
In the project Melody in Human-Cat Communication (Meowsic) we are using established phonetic methods to collect, annotate, pre-process and analyse domestic cat-human vocal communication. This article describes these methods, and also presents results of meow vocalisations in four different mental states showing variation in fundamental frequency (f...
Kulning is a Swedish cattle call singing style with an almost mythical status in Swedish folklore. In previous studies two of the authors (RE, AM) studied kulning produced by a kulning singer (FP) in both indoor and outdoor settings. In this paper we report kulning as produced by a second singer (the third author, KD), recorded outdoors in a forest...
This paper describes a unique singing mode, tentatively labeled "polyphonic overtone singing". In overtone singing the vocal harmonics of a stabile fundamental frequency are filtered by the singer in such a way that specific upper harmonics are amplified, and heard clearly, as a second musical voice. In the "throat singing" of Tuva (Mongolia) movin...
OBS! Denna artikel är publicerad i två versioner: (1) en förkortad version i tryck; (2) en längre version på webben för nedladdning. Denna PDF inkluderar båda versionerna, med anvisningar för hur artikeln skall refereras till.
Segment prolongation has been shown to be one of the most common forms of non-pathological speech disfluency. The distribution in the word (initial–medial–final segment) seems to vary across languages based on morphological complexity, making it interesting to study segment prolongation in languages that exhibit different degrees of morphological c...
Inom akademika ingår som en central del av produktion av vetenskapliga uppsatser, vilket inkluderar formella aspekter av detta – från att lära sig strikt typografiska konventioner som att lära sig hur dessa kan skilja sig åt mellan avdelningar, länder, tidskrifter.
Väldigt ofta, och tyvärr, läggs mycket tid på sådan under såväl handledning som exa...
Within academia a central activity is the production of scientific theses and papers, including formal aspects of this activity – learning about typographical conventions, how these can differ between departments, countries, journals and publishers, etc.
Very often, and sadly, huge amount of of time is spent on this during both thesis supervision...
The recently funded, five-year, project Melody in Human-Cat Communication (Meowsic) has received vast media attention, both nationally and around the world. In this paper we summarize how our activities got started, our published results so far, the present situation and how we envision our planned, future research, including some of the core hypot...
experimental project to study dolphin communicative behaviour using distributional semantics, with methods implemented for the large scale study of human language.
The cat (Felis catus, Linneaus 1758) has lived around or with humans for at least 10,000 years, and is now one of the most popular pets of the world with more than 600 million individuals [1], [2]. Domestic cats have developed a more extensive, variable and complex vocal repertoire than most other members of the Carnivora, which may be explained by...
This paper gives a brief introduction to the starting points of an experimental project to study dolphin communicative behaviour using distributional semantics, with methods implemented for the large scale study of human language.
Segment prolongation (PR) has been shown to be one of the most common forms of non-pathological speech disfluencies (Eklund, 2001). The distribution of PRs in the word (initial–medial–final segment) seems to vary between languages of different syllable-structure complexity, making it interesting to study segment prolongation in languages that exhib...
We investigate segment prolongation as a means of disfluent hesitation in spontaneous German speech. We describe phonetic and structural features of disfluent prolongation and compare it to data of other languages and to non-disfluent prolongations.
There are several studies about non-fluency in people who stutter, but comparatively few regarding children with language impairment. The current research body regarding disfluencies in children with language impairment has been using different study-designs and definitions, making some results rather contradictory. The purpose of the present study...
The Swedish cattle call song, kulning, is an example of very marked and far-reaching sound propagation of vocal communication. While earlier studies have investigated the acoustic characteristics of kulning, the present study focuses on its physiological basis from the point of view of vocal fold function and supralaryngeal posture by applying elec...
Eklund, Robert & Janne Lindberg. 1993.
An Algorithm for End-of-Sentence Detection in Text.
Internal report. Infovox AB, Text-to-Speech Division, Stockholm.
Eklund, Robert & Helén Kåselöv. 1992.
Några observationer rörande akustiskt korrelat till restriktiv bisats.
Bachelor’s Degree paper in Speech Technology (53 pages). Institute of Linguistics, Stockholm University and internal report, Swedish Telecom.
Linguistics paper (101) on first name connotations.
Corrected version of PhD thesis in Computational Linguistics. Disfluency in Swedish human–human and human–machine travel booking dialogues.
PhD thesis, Linköping Studies in Science and Technology, Dissertation No. 882, Department of Computer and Information Science, Linköping University, Sweden, ISBN 91-7373-966-9, ISSN 0345-7524.
Course compendium in Prolog programming. [In Swedish.]
A short introdution to machine translation. [In Swedish.]
Crib sheet on syntactic functions and categories
A conference paper based on my BA thesis in Computational Linguistics. NODALIDA ’93 – Proceedings of ‘9:e Nordiska Datalingvistikdagarna’,
Stockholm 3–5 June 1993. ISBN 91-7153-262-5, pp. 83–95.
A short workshop paper on university pedagogics [In Swedish.]: Britt Rönnbäck (ed.): Vad gör vi för att förbättra kvalitén på utbildningen? Dokumentation från campuskonferens vid Stockholms universitet 15 november, 1994, s 13-14, Institutionen för Pedagogik, Stockholms Universitet.
A short description of the Universal Turing Machine (1993). [In Swedish.]
BA thesis in Computational Linguistics (1993) on automatic tagging.
The charlatan of music. A history of the lute. [In Swedish]Tidskrift för Tidig Musik. Vinternummer. 1992. Nr 4, årgång 14, s 5–13. (Article.)
Tidskrift för Tidig Musik. Vinternummer. 1993. Nr 1, årgång 15, s 10. (Correction.)
Notification in Svensk Tidskrift för Musikforskning (volume 73, 1991, Göteborg 1992, p. 122.) on my MA thesis.
MA thesis musicology on a unique, late baroque lute manuscript.
A short paper on 18th century Swedish lutenists.
An introduction to editing practices of early music (medieval, renaissance, baroque) into modern notation. In Swedish.
The Swedish cattle call singing style ‘kulning’ is
surprisingly understudied, despite its almost
mythical status in Swedish folklore. While some
physiological-productive aspects of kulning have
been treated in previous work, acoustic properties
are still much lacking description. This paper adds
to and extends the results presented in a previous
st...
This paper reports the prevalence of disfluencies in a
group of 55 (25F/30M) Swedish children with
typical speech development, and within the age
range 6;0 and 6;11. All children had Swedish as
their mother tongue. Speech was elicited using an
“event picture” which the children described in their
own, spontaneously produced, words. The data were
an...
Spontaneously produced Unfilled Pauses (UPs) and Filled Pauses (FPs) were played to subjects in an fMRI experiment. While both stimuli resulted in increased activity in the Primary Auditory Cortex, FPs, unlike UPs, also elicited modulation in the Supplementary Motor Area, Brodmann Area 6. This observation provides neurocognitive confirmation of the...
Recent years have seen a growing number of studies on both felid vocalizations in general and human–felid communication in particular. Frequently considered as the starting point for this line of research is Mildred Moelk's seminal paper from 1944, in which she provides a taxonomy of basic felid vocalizations, complete with phonetic transcriptions....
Speaking on inhalation, pulmonic ingressive speech, is well-known in Scandinavia and often believed to be unique to this part of the world. It has, however been shown (Eklund, 2002, 2007, 2008) that not only is ingressive speech not confined to the northernmost part of Europe, it is found all over the world and might be regarded as a linguistic uni...
This paper summarizes recent research on 'kulning', a surprisingly understudied Swedish cattle call singing style. In a previous study (Eklund, McAllister & Pehrson, 2013), we compared kulning and head voice ('falsetto') as recorded in a normal room and in an anechoic chamber. This paper reports from an analysis of the same " kulning " song recorde...
Full conference proceedings, edited by Robert Eklund.
ISBN 978-91-981276-0-7
eISBN 978-91-981276-1-4
ISSN 1104-5787
ISRN KTH/CSC/TMH--13/01-SE
TRITA TMH 2013:1
Previous studies of cheetah purring have described purring in adult cheetahs. This paper extends the cheetah purring research to include juvenile and subadult cheetahs and analyzes purring data from cheetahs in ages ranging from 7 months to 7 years, and with weights ranging from 18 kilos to over 70 kilos. Results show that while there is considerab...
Proceedings of Fonetik 2013
The XXVIth Annual Phonetics Meeting
12–13 June 2013, Linköping University, Linköping, Sweden
Studies in Language and Culture, no. 21
Robert Eklund, editor
ISBN 978-91-7519-582-7
eISBN 978-91-7519-579-7
ISSN 1403-2570
We report results from a longitudinal study of the rate and location of disfluencies in child-directed speech, using data for children between 0;6 and 2;9 years. We compare these results to adult-directed speech by the same speakers.
The Swedish folk singing style 'kulning' is surprisingly understudied, despite its almost mythical status in Swedish folklore. While some physiological-productive aspects of kulning have been treated in previous work, acoustic properties are still much lacking description. This paper compares kulning, head ('falsetto') and modal voice from an acous...
We describe a Swedish version of CALL-SLT, a web-deployed CALL system that allows beginner/intermediate students to practise generative spoken language skills. Speech recognition is grammar-based, with language models derived, using the Regulus platform, from substantial domain-independent feature grammars. The paper focusses on the Swedish grammar...
This paper reports results from a comparative analysis of purring in four domestic cats. An acoustic analysis describes sound pressure level, duration, number of cycles and fundamental frequency for egressive and ingressive phases. Significant individual differences are found between the four cats in several respects.
This paper studies the frequency and distribution of filled pauses (FPs) in ecologically valid data where unaware and authentic customers called in to report problems with their telephony and/or Internet services and were met by a novel Wizard-of-Oz paradigm using real call center agents as wizards. The data analyzed were caller utterances followin...
This paper looks at the phenomenon of ingressive speech, i.e. speech produced on a pulmonic ingressive airstream, set in the context of human and animal ingressive phonation. The literature on ingressive speech and phonation spanning several centuries is reviewed, as well as contemporary reports of their incidence and characteristics from both func...
This paper describes our experiences of collecting a corpus of 42,000 dialogues for a call-routing application using a Wizard-of-Oz approach. Contrary to common practice in the industry, we did not use the kind of automated application that elicits some speech from the customers and then sends all of them to the same destination, such as the existi...
This paper summarizes major review work on pulmonic ingressive speech (Eklund, under revision), e.g., words like ja (yes) and nej (no) that are commonly produced on inhalation airstream in Swedish. Contrary to what is generally believed, ingressive speech is not limited to Scandinavia or present-day Nordic languages. Instead, it is shown that ingre...
The conformational preference of α-l-Rhap-(1→2)[α-l-Rhap-(1→3)]-α-l-Rhap-OMe in solution has been studied by NMR spectroscopy using one-dimensional 1H,1H T-ROESY experiments and measurement of trans-glycosidic 3JC,H coupling constants. Molecular dynamics (MD) simulations with a CHARMM22 type of force field modified for carbohydrates were performed...
Proceedings of DiSS’03 – Disfluency in Spontaneous Speech.
Robert Eklund, editor.
ISSN 0349-1021.
In this paper, we compare the distribution of disfluencies in two human--computer dialogue corpora. One corpus consists of unimodal travel booking dialogues, which were recorded over the telephone. In this unimodal system, all components except the speech recognition were authentic. The other corpus was collected using a semi-simulated multi-modal...
Languages have always been influenced by other languages in various ways, through cultural contacts, migration, trade and other channels. In an increasingly internationalized world, where contacts across national borders are commonplace, sometimes politically driven/pushed by bodies such as the EU, foreign language influences have become stronger t...
In recent years, both automatic speech recognition (ASR) and text-to-speech (TTS) conversion systems have attained quality levels that allow inclusion in everyday applications. One remaining problem to be solved in both these types of applications is that alleged phone inventories of specific languages are commonly expanded with phones from other l...
This paper studies disfluencies in authentic human-human dialogues in Swedish and Tok Pisin. It is found that while there are no major differences as to types or frequencies on a macro level, there are dissimilarities on a micro level, notably in the characteristics of how prolonged segments are realized. The paper also discusses the results in the...
An abstract is not available.
In this paper, the distribution of Swedish subjects’ productions
of foreign speech sounds, here termed xenophones, is studied,
and tabulated across gender, age, and region. The results are
grouped in three categories along the “awareness” and “fidelity”
dimensions. Results indicate that age is by far the most decisive
underlying factor, which can b...
This paper discusses the problem of handling "foreign" speech sounds in Swedish speech technology systems, in particular speech synthesis. A production study is made, where it is shown that Swedish speakers add foreign speech sounds, here termed 'xenophones', to their phone repertoire when reading Swedish sentences with embedded English names and w...
This paper deals with the treatment of foreign words and proper names in Swedish. Preliminary results from a production study are presented, and guidelines are suggested for broad, phonematic transcription, covering alternative pronunciations. Such a transcription scheme is a prerequisite for applications such as speech synthesis and multi-dialecta...
This paper describes an operational speech-to-speech translation system from Swedish to Tok Pisin within the framework of the Spoken Language Translator project, SLT [1]. The domain of translation is ATIS [11]. The grammar formalism used in the SLT project is the Core Language Engine, CLE [2]. A general presentation of Tok Pisin is provided, as wel...
Automatic speech understanding systems are beginning to attain a level of sophistication where commercial applications are within reach. However, if humans and machines are ever going to communicate in a natural way, it is of vital importance that language modeling go beyond the sentence level. A profound understanding of discourse structure is req...
Experiments have been conducted that deal with prosodic prominence in reiterant speech in order to determine the relative contribution of F0 and duration to the perception of prosodic prominence by Swedish listeners. F0 and duration were manipulated independently on different syllables in the stimuli. The results show that F0 is considered primary...
In this paper we describe how the translation methodology adopted for the Spoken Language Translator (SLT) addresses the characteristics of the speech translation task in a context where it is essential to achieve easy customization to new languages and new domains. We then discuss the issues that arise in any attempt to evaluate a speech translato...
We describe two methods relevant to multi-lingual machine translation systems, which can be used to port linguistic data (grammars, lexicons and transfer rules) between systems used for processing related languages. The methods are fully implemented within the Spoken Language Translator system, and were used to create versions of the system for two...
State-of-the-art speech recognition systems handle continuous speech and are speaker-independent. However, the linguistic information conveyed in the intonational contour is neglected. To be able to fully recognize speech, this information must be interpreted. To this end, explicit knowledge of dialectal and individual variation is required. In thi...
State-of-the-art speech recognition systems handle continuous
speech and are speaker-independent. However, the linguistic information
conveyed in the intonational contour is neglected. To be able to fully
recognize speech, this information must be interpreted. To this end,
explicit knowledge of dialectal and individual variation is required.
Some a...
The Spoken Language Translator (SLT) is a multi-lingual
speech-to-speech translation prototype supporting English, Swedish and
French within the air traffic information system (ATIS) domain. The
design of SLT is characterized by a strongly corpus-driven approach,
which accentuates the need for cost-efficient collection procedures to
obtain training...
State-of-the-art speech recognition and speech translation systems do not currently make use of prosodic information. Utterances often have one or more constituents semantically focused by prosodic means and detection of the focus/foci of an utterance is crucial for a correct interpretation of the speech signal. Thus, a semantic model of focus shou...