Urdu Pitch Accents and Intonation Patterns in Spontaneous Conversational Speech

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 6412–6416 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 6412 Urdu Pitch Accents and Intonation Patterns in Spontaneous Conversational Speech Luca Rognoni, Judith Bishop, Miriam Corris, Jessica Fernando, Rosanna Smith Appen Limited, Australia Level 6, 9 Help Street Chatswood NSW 2067, Australia {lrognoni, jbishop, mcorris, jfernando, rsmith}@appen.com Abstract An intonational inventory of Urdu for spontaneous conversational speech is determined based on the analysis of a hand- labelled data set of telephone conversations. An inventory of Urdu pitch accents and the basic Urdu intonation patterns observed in the data are summarised and presented using a simplified version of the Rhythm and Pitch (RaP) labelling system. The relation between pitch accents and parts of speech (PoS) is also explored. The data confirm the important role played by low pitch accents in Urdu spontaneous speech, in line with previous studies on Urdu/Hindi scripted speech. Typical pitch contours such as falling tone in statements and WH-questions, and rising tone for yes/no questions are also exhibited. Pitch accent distribution is quite free in Urdu, but the data indicate a stronger association of pitch accent with some PoS categories of content word (e.g. Nouns) when compared with function words and semantically lighter PoS categories (such as Light Verbs). Contrastive focus is realised by an L*+H accent with a relatively large pitch excursion for the +H tone, and longer duration of the stressed syllable. The data suggest that post-focus compression (PFC) is used in Urdu as a focus-marking strategy. Keywords: Urdu, intonation, prosody annotation, pitch contour, RaP, ToBI, PFC, part of speech 1. Introduction Recent studies of Urdu intonation describe scripted and laboratory speech, using ToBI-like annotations to characterise Urdu pitch accents and intonation patterns (Nayyer and Madni, 2002; Jabeen, 2010). The novelty of the data presented here is that they contain spontaneous conversational speech rather than read or scripted material. Speech recorded in this manner (unscripted telephone conversations between friends or relatives) is likely to represent better the intonation patterns of ordinary speech. It has been argued that, based on the structural similarity between Urdu and Hindi, intonation patterns and prosodic marking strategies observed in Hindi can also be expected in Urdu (Masica, 1991; Dyrud, 2001; Nyyar, 2002; Hock and Bashir, 2016). Sengar and Mannel (2012) have proposed L*+H as the “default pitch accent” for Hindi, in agreement with Patil et al. (2008), who suggest the p-phrase LH as the main prosodic building block of Hindi. The data presented by Nayyer and Madni (2002) and Jabeen (2010) suggest that this could also apply to Urdu. The latter two studies specifically analyse Urdu intonation patterns in scripted speech. They confirm that statements and WH-questions (WHQ) are realised with a falling intonation, while yes/no Questions (YNQ) are characterised by a rising contour. Studies of the acoustic- phonetic realisation of word stress in Urdu show that this is often aligned with a low tone (L*) (Hussain (1997); Dyrud, 2001). In a …

Content truncated to 500 words