International Speech Communication Association (Isca) - International Astronautical Federation. ISBN : 13 9781604234480.

We present the results of a large-scale statistical study of pitch target points in five languages, on a corpus comprising 4 hours 20 minutes of speech and involving 50 different speakers. The entire corpus has been stylized automatically by a technique reducing the F 0 contour to a series of target points representing the significant pitch changes. It was then entirely verified by experts using a resynthesis method, in order to ensure that there was no audible difference with the original. The set of ca. 50000 pitch target points thus obtained was then analyzed from a statistical point of view. In this paper we describe the main results of this study, in terms of frequency distribution of target points, pitch movements and relation of pitch movements to time interval. Our study reveals interesting differences across languages and sex. 1. INTRODUCTION Large sets of prosodic data on many languages would be a useful resource for theoretical studies, as well as for practical application...

We present a prosodic corpus in five languages (French, English, Italian, German and Spanish) comprising 4 hours and 20 minutes of speech and involving 50 different speakers (5 male and 5 female per language). The recordings on which the corpus is based are extracted from the EUROM 1 speech database and consists of passages of about five sentences. The corpus was stylized automatically by an algorithm which factors out microprosodic effects and represents the intonation contour of utterances by a series of target points. Once interpolated by a smooth curve (spline), these points produce a contour undistinguishable from the original when re-synthesized, apart from a few detection errors. A symbolic coding of the 50000 pitch movements of the corpus is also provided, along with the time-alignment of orthographic transcription to signal at wordlevel. The entire corpus was verified and manually corrected by experts for each language. It will be made available at production cost for research...

This paper presents a two-step model for the symbolic coding and generation of intonation. First, the F 0 curve is reduced to a series of pitch target points that capture the macroprosodic information of the utterance. Target points are then converted into a sequence of labels. Generation is achieved through the reverse steps. The model is language independent and requires no prior training on the data. We discuss the influence of the number of categories on the precision of fit, and show, by an evaluation on a large multilingual corpus (4 hours 20 minutes of speech, 50 speakers, 5 languages) that a model composed of three ascending and three descending categories, plus a category for small or null movements enables a regeneration of ca. 99% of points at less than 2 ST than the original. Given that the model is capable of various improvements, it seems a good candidate for practical applications. 1. INTRODUCTION Several systems of symbolic coding of intonation have been proposed, but...

This paper presents a stochastic model of French intonation contours for use in text-to-speech synthesis. The model has two modules, a linguistic module that generates abstract prosodic labels from text, and a phonetic module that generates an F 0 curve from the abstract prosodic labels. This model differs from previous work in the abstract prosodic labels used, which can be automatically derived from the training corpus. This feature makes it possible to use large corpora or several corpora of different speech styles, in addition to making it easy to adapt to new languages. The present paper focuses on the linguistic module, which does not require full syntactic analysis of the text but simply relies on a part-of-speech tagging technique. The results were validated by means of a perception test which showed that listeners did not perceive a significant difference in quality between the sentences synthesized with the original F 0 curve (from a recording), and those synthesized with the...

In this paper, we point out that, in applications available to the general public, and/or natural language interfaces, the correction of phonographic errors (which are competence errors) is far more important than the correction of typographical errors (which are simply performance errors). Many studies aimed at the correction of typographical errors have been carried out, but relatively few tackle the problem of phonographic correction, and they are generally based on more or less ad hoc methods. We propose a mathematical framework for phonographic correction by defining a similarity relation between phonetically related substrings and a dissimilarity index between strings. We also provide a simple and efficient algorithm for recognizing words in dictionaries from misspelt inputs including both typographical and phonographic errors.