Lennart Nord

Temporal Speech Characteristics of Individuals with Multiple Sclerosis and Ataxic Dysarthria: ‘Scanning Speech’ Revisited

‘Scanning speech’ has been used as a description of a prominent characteristic of the dysarthria of multiple sclerosis (MS) as well as of ataxic dysarthria in general. It is thought to be measurable as equalised syllable durations. There are seemingly contradictory prosodic-temporal characteristics of ataxic dysarthria: perceptually characterised as prosodic excess as well as phonatory-prosodic insufficiency and acoustic characteristics including signs of isochrony as well as variability. This study investigates the temporal characteristics at two levels, duration and variability of syllable durations and the durations of interstress intervals. A group of 14 individuals with MS and ataxic dysarthria as well as 15 control subjects were studied. It was concluded that individuals with ataxic dysarthria and MS showed (a) for syllables: significantly increased durations and decreased intrautterance variability (more isochrony or syllable equalisation) as well as significantly increased interutterance variability; (b) for interstress intervals: significantly increased durations and increased variability (less isochrony). The results point to inflexibility as well as instability of temporal control, which could contribute to the explanation of why the perceptual characteristics are contradictory.

Prosodic and segmental speaker variations

Abstract The purpose of our presentation is to make an inventory of knowledge developed at the KTH about speaker variabilities including findings from our more recent databank projects on text reading. We shall have something to say about male/female differences, voice source characteristics, and about prosodic and segmental features in connected speech. We also have some data of more general statistical nature such as pause durations, long time average spectrum, and about relative proportions of voiced and voiceless segments in speech.

Acoustic analysis of dysarthria associated with multiple sclerosis

Approximately 40% of all individuals with multiple sclerosis (MS) present some degree of speech impairment. MS speakers with dysarthria are a heterogeneous group and the speech disorder is characterized by disturbances of the temporal patterning of speech as well as articulatory and phonatory symptoms. Previous research also indicates that the neuromotor dysfunction manifests itself in the fundamental frequency of sustained phonation, even in the absence of other speech symptoms. The present study was designed with the specific purpose of exploring the temporal, spectral and phonatory acoustic features of five MS speakers as compared to those of two normal controls. It was found that the dysarthric symptoms of these MS speakers largely mirror their different underlying neuromotor dysfunctions, and that they exhibit temporal and articulatory deviation when compared to the normal speakers. It is also suggested that a Fourier spectrum analysis of the derived pitch of the sustained phonation in MS speakers is...

Neglected dimensions in speech synthesis

Abstract In traditional accounts on speech prosody, fundamental frequency, duration and intensity have been described as the most important attributes. Among these, intensity has attracted the least attention.In perceptual studies both F0 and duration have had an undisputable role in signalling prosodic categories, but the role of intensity has been less clear. This has resulted in an emphasis on the former attributes in current speech synthesis schemes. We are in this study exploring the use of speech intensity and also other segmental correlates of prosody. Intensity has a dynamic aspect, discriminating emphasized and reduced stretches of speech. A more global aspect of intensity must be controlled when we try to model different speaking styles. Specifically, we have been trying to model the continuum from soft to loud speech.

Evaluation and development of the KTH text-to-speech system on the segmental level

A procedure that has been used for several years in developing the Swedish version of the KTH multilingual text-to-speech system on the segmental level is described. Experimental results from an evaluation of synthetic speech by 12 untrained listeners are presented. A method for improved nasal synthesis is discussed.<<ETX>>

The KTH speech database

Abstract In current speech research, there is a need for large databases to be able to test production and perception models at different linguistic levels. There are considerable problems in administering databases, both to label the speech and to easily access stored material. In order to alleviate some of the problems we have created a speech analysis system. Speech data are stored in sentence-sized files. These files are segmented and transcribed semi-automatically given a phonetic transcription of the utterance. This transcription is generated by the letter-to-sound rules of our text-to-speech system. The emphasis on the database is the use for acoustic-phonetic research rather than the use in e.g. evaluation of speech recognizers. This makes demands on flexible and linguistically specified retrieval patterns. Our unorthodox solution to this is to use the synthesis rule structure, similar to the notation used in generative phonology, for accessing the data. By a brief rule statement, speech segments meeting the specified contextual conditions can be identified. Durational data can be collected directly during the database search. Spectral analysis programs operating with a variety of spectral representations have also been created that display the result, typically as a mean/standard deviation spectrum or as a contour histogram spectrum.

Laryngectomee speech in noise—voice effort, speech rate and intelligibility

Different aspects of alaryngeal speech, both esophageal and tracheo-esophageal speech, are being analysed in a joint project between the Department of Speech Communication and Music Acoustics, KTH, and the Department of Logopedics and Phoniatrics, Karolinska Institute. The purpose of the present part of the project was to evaluate the speech performance of laryngectomee speakers while they were reading texts aloud with varying amounts of noise in their ears. The noise consisted of a number of voices in a cacophony. Subjects were four men, two of whom were esophageal speakers, and two tracheo-esophageal speakers. Acoustic speech parameters, such as sound pressure and spectral characteristics, were measured and compared among the subjects. The results showed that the tracheo-esophageal speakers were able to raise their voice level almost as much as normally speaking subjects. The esophageal speakers on the other hand were usually not able to produce as strong voice levels during the text readings. This type...

Some timing studies of prose, poetry and music

Abstract Prosodic relations in prose, poetry and music are discussed with an emphasis on durational properties. In order to gain a deeper understanding of speech prosody, we are presently engaged in a comparison of the timing relations in such activities as the reading of poetry and music performance, where there usually is a strong and obvious rhythmic patterning of the produced sound sequences. Also there are interesting parallels to be drawn by comparing the formal notations of prose, poetry and music. Generally, there are no simple relations between abstract notations and performance, and moreover, notations have varied with tradition and particular needs. However, it is a challenge to tie descriptive systems closer to common human constraints in production and perception.

Does ‘‘one hundred fifty‐five’’ mean 155 or 100,55 or 100,50,5?

The goal of this study is to determine what are the acoustic cues, if any, that differentiate the spoken phrase that means 155 from one that means 100,55 or others for 150,5 or 1,155 or 100,50,5. The motivation for this study comes from an issue that arose in designing a speech‐user interface—that is, a user interface controlled by a user’s speech via a speech recognizer. Users of a speech‐user interface for programs that require numerical input (engineering design systems, accounting packages, etc.) need to speak sequences of numbers such as 100,50,5 (for example, to specify a three‐dimensional coordinate location) as well as 100,55 (for a two‐dimensional coordinate location). In this study, speakers produced utterances of one‐, two‐, and three‐number sequences in two languages: Swedish and English. These productions were analyzed acoustically in terms of fundamental frequency, syllable duration, and amplitude. Listeners identified for each production whether the speaker intended to specify one, two, or ...

Data-Bank Analysis of Speech Prosody

Publisher Summary This chapter discusses the analysis procedures and results from studies of speech prosody and individual variations in text reading. The chapter aims at deriving rules for good reading performance and rules related to voice types, sex, age, and reading style. This chapter discusses segmentation problems and provides data on objective and subjective studies of syllabic stress, speech rhythm, and the realization of phrase and sentence boundaries. This chapter explains a provisional phonemic transcription and segmentation, which conforms to the letter-to-sound rules of the RULSYS synthesis. Phonemes may be so weakly manifested and be subjected to such extreme temporal spread that signal-driven segmentation strategies fail.