Search

Phonetik und Phonologie im deutschsprachigen Raum, Phonetik und Phonologie im deutschsprachigen Raum

International audience; Work done in recent years has shown the usefulness of using automatic methods for the study of linguistic typology. However, the majority of proposed approaches come from natural language processing and require expert knowledge to predict typological information for new languages. An alternative would be to use speech-based methods that do not need extensive linguistic annotations, but considerably less work has been done in this direction. The current study aims to reduce this gap, by investigating a promising speech representation, i-vectors, which by capturing suprasegmental features of language, can be used for the automatic characterization of languages. Employing data from 24 languages, covering several linguistic families, we computed the i-vectors corresponding to each sentence and we represented the languages by their centroid i-vector. Analyzing the distance between the language centroids and phonological, inventory and syntactic distances between the same languages, we observed a significant correlation between the i-vector distance and the syntactic distance. Then, we explored in more detailed a number of syntactic features and we proposed a method for predicting the value of the most promising feature, based on the i-vector information. The obtained results, an 87% classification accuracy, are encouraging and we envision to extend this method further.

Synthetic speech has a long standing tradition of being employed for experiments in phonetics and laboratory phonology. The choice of synthesis method and system is commonly made by the researcher(s) to fit the specific quality criteria and study design. The overall quality of a given system, however, remains as a confound that is difficult to control for [1]. In speech technology newly proposed systems are usually compared across specific dimensions e.g., ‘Intelligibility’ and ‘Naturalness’. These dimensions have already been extensively studied and evaluated within the context of old diphone and formant synthesis networks [2]. We contend, however, that these tradi- tional dimensions need to be re-examined in the context of state of the Art Text-to-Speech (TTS) systems, as those newer models exhibit different quality deteriorations. Our work aims to bridge the conflicting demands for quality criteria that are easily computed and applied during TTS de- velopment, while at the same time remaining descriptive and meaningful for phonetic research. As a first step in this endeavor, we carried out an experiment to find suitable dimensions of TTS quality with a bottom-up approach based on descriptions provided by 11 participants (phonetic experts). The participants were instructed to label speech samples generated by 8 different state of the art Text-to-speech systems (varieties of English). Each system produced a stimulus consisting of two sentences of the phonetically balanced ‘caterpillar story’ [3]. In order to ensure that all systems were evaluated across different phonetic contexts in a balanced way, the sentences were rotated between participants so that each participant heard the complete story but with different parts read by different systems. The experimental setup is loosely based on the work in [4]. The participants were instructed to write down nouns, adjectives or sentences describing the quality of a given stimulus. Using embeddings generated by a pretrained BERT model [5] for semantic distances, we determined which of the participants terms were semantically similar. A subsequent affinity propagation clustering revealed there to be 39 meaningfully different clusters, each rep- resenting a dimension of quality for synthetic voices. Keeping in mind that these dimensions are later to be used for ratings in actual evaluation experiments, it was decided to reduce the num- ber of clusters to a more practical number of 10 and re-calculate the spectral clustering with a precomputed cosine affinity matrix. The resulting clusters and their respective quality descrip- tions are depicted in fig. 1. A manual analysis of the resulting dimensions led to the following descriptive labels: ‘artificiality/voice quality’, ‘intonation/noise/prosody’, ‘voice/audio quality’, ‘audio cuts’, ‘style/recording quality’, ‘emotion/voice quality/attitude’, ‘engagedness’, ‘human likeness’, ‘hyperarticulation’. From the assigned cluster descriptions it is evident, that the se- mantic embeddings sometimes conflated several seemingly unrelated quality features into single dimensions (e.g. prosody and background noise), while occasionally splitting almost synonymous terms into multiple clusters (e.g. ‘artificiality’, ‘roboticness’ and ‘metallicness’). To evaluate these shortcomings of the semantic model, two independent manual clusterings were carried out. They were both limited to 10 clusters and reported a modified jaccard agreement index of 63,44, while agreeing with the automatic computed clusters with 54.48 and 57.93, respectively. The low in- terrater agreement between the manual clusters suggests that a panel decision process might be needed to determine the final quality dimensions. Subsequent research will evaluate clusters cre- ated by na ̈ıve listeners and quality dimensions of different sub-tasks in synthetic speech, such as voice conversion.

This study aims to quantify the effect of several information sources: acoustic, higher-level linguistic, and knowledge of the prosodic system of the language, on the perception of prosodic boundaries. An experiment with native and non-native participants investigating the identification of prosodic boundaries in Japanese was conducted. It revealed that non-native speakers as well as native speakers with access only to acoustic information can recognize boundaries better than chance level. However, knowledge of both the prosodic system and of higher-level information are required for a good boundary identification, each one having similar or higher importance than that of acoustic information.& nbsp;(C) 2021 Author(s). All article content, exceptwhere otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

We present here a study on the use of laughter in spontaneous interactions, examining its distribution across two linguistic levels: utterances and turns. A multilingual corpus of dyadic conversations was employed, containing recordings in French, German and Mandarin Chinese. Laughter was coded based on its position inside the analysis unit and its distribution with respect to the event type and the language was analyzed. The results showed that laughter distribution is modulated by the linguistic level, as well as by the laughter event type. Moreover, differences between languages seem to depend on the analysis level., Laughter and Other Non-Verbal Vocalisations Workshop: Proceedings (2020), Laughter and Other Non-​Verbal Vocalisations Workshop 2020

With laughter research seeing a development in recent years, there is also an increased need in materials having laughter annotations. We examine in this study how one can leverage existing spontaneous speech resources to this goal. We first analyze the process of manual laughter annotation in corpora, by establishing two important parameters of the process: the amount of time required and its inter-rater reliability. Next, we propose a novel semi-automatic tool for laughter annotation, based on a signal-based representation of speech rhythm. We test both annotation approaches on the same recordings, containing German dyadic spontaneous interactions, and employing a larger pool of annotators than previously done. We then compare and discuss the obtained results based on the two aforementioned parameters, highlighting the benefits and costs associated to each approach.

Laughter is a paralinguistic phenomenon widely used in human communication. Previous studies on laughter have mainly looked at its acoustic realization and its functions, leaving the context in which laughter occurs relatively under-studied. We intend to partially fill this gap by conducting an investigation into the acoustic cues that mark the use of laughter. We focus on the syllables preceding laughter and we explore several relevant spectral features. The results obtained on an American English corpus of conversational speech show anticipatory effects on the syllable immediately preceding laughter: a higher F1, a higher spectral center of gravity and a greater spectral standard deviation. We discuss these findings in terms of the individual variation present in laughter.

Prosodic prominence is an umbrella term encompassing various related but conceptually and functionally different phenomena such as phonological stress, paralinguistic emphasis, lexical, syntactic, semantic or pragmatic salience, to mention a few. Due to the high interest prominence has received from various disciplines, it has been studied from multiple perspectives (functional, physical, cognitive). It also has been operationalised and annotated across different descriptive levels (syllable, word), based on different scales (categorical, multi-level, continuous), and measured across a large variety of signal domains (acoustic, articulatory, gestural). The present paper offers an overview of the various perspectives involved and defines a preliminary roadmap for a better and more unified understanding of this multi-faceted phenomenon.

International audience; This work presents a novel framework to guide the Viterbi decoding process of a hidden Markov model based speech recognition system by means of broad phonetic classes. In a first step, decision trees are employed, along with frame and segment based attributes, in order to detect broad phonetic classes in the speech signal. Then, the detected phonetic classes are used to reinforce paths in the search process, either at every frame or at phonetically significant landmarks. Results obtained on French broadcast news data show a relative improvement in word error rate of about 2% with respect to the baseline.

National audience; The main aim of this work is to provide a set of tools for automatic segmentation of Italian speech into phones. We make available already trained acoustic models, a compiled version of the automatic aligner, a script to convert the segmentation into PRAAT format and the necessary documentation. The acoustic models have been trained with the SPRAAK toolkit, using HMMs extracted from the CLIPS corpus. In order to evaluate performances, first of all we measured phone markers distances between our segmentation and a reference one, using both a test corpus and the APASCI database. Then we also compared the results of metrics applied to manual and automatic segmentations in the frame of an independent study on Neapolitan Italian prosody. Our results show that the automatic alignment provided by ASSI is a fast and reliable substitute for manual segmentation in lab-speech based linguistic studies.