Santiago Arróniz

This paper examines the effectiveness of different feature representations of audio data in accurately classifying discourse meaning in Spanish. The task involves determining whether an utterance is a declarative sentence, an interrogative, an imperative, etc. We explore how pitch contour can be represented for a discourse-meaning classification task, employing three different audio features: MFCCs, Mel-scale spectrograms, and chromagrams. We also determine if utilizing means is more effective in representing the speech signal, given the large number of coefficients produced during the feature extraction process. Finally, we evaluate whether these feature representation techniques are sensitive to speaker information. Our results show that a recurrent neural network architecture in conjunction with all three feature sets yields the best results for the task.

This study examines how variable intonational contours are (mis)interpreted in two different Spanish dialects, Western Andalusian (Henriksen and Amaya 2012) and Puerto Rican Spanish (Armstrong 2010). We seek to investigate how listeners from both dialects interpret the sentence meaning of declaratives, broad focus interrogatives and counter-expectational interrogatives using pitch as the only phonological cue. To do that, two speakers from Puerto Rico and one speaker from Seville, Spain, produced the stimuli for the different perception experiments based on the contours documented in the Interactive Atlas of Spanish Intonation (Prieto and Roseano 2010). An utterance type selection task and a declarative to interrogative continuum judgment task were created from these utterances. Both tasks were presented through the Qualtrics platform to 57 Puerto Rican listeners and 35 Andalusian listeners. The findings indicated that both groups could identify declarative contours as declaratives. Both interrogative types of Puerto Rican contours were consistently identified as interrogatives by the Puerto Ricans listeners. Andalusian listeners were less certain recognizing broad focus interrogatives as a question. Counter-expectational contours were recognized as declaratives by Andalusian listeners. Results indicate that specific intonation patterns can be interpreted differently based on Spanish dialect and may result in the misinterpretation of the utterance.

This study aims to investigate the acquisition of trills in heterosyllabic sequences by speakers of Spanish as a second language (L2) and heritage speakers. To this end, we presented participants with three oral production tasks designed to elicit instances of trills, accompanied by a motivation questionnaire and a background questionnaire. The results obtained by means of acoustic analysis of the productions together with the answers to the motivation questionnaire show clear acquisitional patterns in L2 learners of Spanish at varying levels of proficiency, as well as variation in native speakers’ production of the trill in heterosyllabic sequences. This paper also raises some pedagogical implications, pointing out that instructors should more explicitly teach and illustrate examples of trills in heterosyllabic sequences at all levels, since this phenomenon is neither orthographically salient nor frequent in speech.

This work undertakes an empirical study of one of the least researched issues in interlanguage pragmatics so far: comprehension of jokes by Spanish learners of English as a second language (L2, henceforth). It seeks to examine whether these learners grasp the humor some joking texts are intended to produce, which types of jokes prove to be more difficult to understand and why. Adopting a relevance-theoretic perspective on comprehension, and, more precisely, the model of comprehension envisaged in the cognitive-pragmatic framework put forward by Sperber and Wilson (1986/1995), this work relies on the classification of jokes recently proposed by Yus (2016), where jokes are sorted out into seven different types depending on three parameters which are purportedly exploited by humorists in their creation and are, therefore, responsible for their triggering the expected effect. Consequently, this work presents the results obtained through an online questionnaire that was administered to students of English as an L2 with an upperintermediate, B2 level who were enrolled in one of the compulsory courses of the Degree in English Studies taught at the Universidad de Sevilla. Owing to its limitations, this work cannot but be considered an initial, exploratory study that aims to spark off further research contributing to a better and deeper understanding of the problems that L2 learners may face when processing humorous texts.

This paper reports on an exploratory study examining joke identification, appreciation and comprehension by Spanish intermediate ESL learners. The study is based on a relevance-theoretic classification of jokes, which assumes that humorousness results from manipulation of three parametres: make-sense frames, cultural information and utterance interpretation. It firstly ascertains whether Spanish ESL learners recognise orally-delivered samples of seven types of purportedly jocular texts. Secondly, it examines whether these learners actually regard such texts as comical and why. Finally, it looks into the learners’ interpretative problems in order to single out which joke type(s) is/are more challenging. The study relies on quantitative and qualitative data elicited through an online questionnaire comprising four tasks. The results indicate no correlation between joke identification and appreciation, and independence of successful joke recognition from sophisticated interpretative skills. Jokes involving invalidation of an activated make-sense frame were most easily identified and found most funny, but jokes exploiting cancellation of an initial, seemingly plausible, interpretation posed more difficulties.