Prosodic annotation in the new corpus of Russian spontaneous speech CoRuSS

Volskaya, Nina B.; Kachkovskaia, Tatiana

doi:10.21437/speechprosody.2016-188

Cited by 11 publications

(15 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Our results on average IP size are roughly in line with the work by Volskaya and Kachkovskaia (2016), who found the average length of IPs to be 3.58 lexical words, or 2.31 prosodic words in CoRuSS. In order to test for statistically significant differences between speaker groups and formality, we ran a linear mixed model in Python with words/IP as the dependent variable and speaker group (HS/monolingual) and formality (formal/informal) as independent variables, also testing the interaction between the variables of speaker group and formality.…”

Section: Intonation Phrasessupporting

confidence: 91%

“…CoRuSS (Kachkovskaia et al 2016;Volskaya and Kachkovskaia 2016) is a corpus of 14 h of Russian spontaneous speech by 60 speakers of three age groups, balanced for gender. Interactive dialogues on free topics were manually transcribed for their prosody, differentiating between 13 basic melodic types (largely following the intonational constructions postulated in Bryzgunova 1980, see Section 1.1).…”

Section: Corpus Studies On Russian Intonationmentioning

confidence: 99%

“…The prosodic annotations of the corpus include automatically detected turn boundaries and inter-pausal units. Further prosodic annotations are partially available (following the approach by Volskaya and Kachkovskaia 2016).…”

Section: Corpus Studies On Russian Intonationmentioning

confidence: 99%

See 2 more Smart Citations

Intonational Features of Spontaneous Narrations in Monolingual and Heritage Russian in the U.S.—An Exploration of the RUEG Corpus

Zerbian,

Zuban,

Klotz

2023

Languages

View full text Add to dashboard Cite

This article presents RuPro, a new corpus resource of prosodically annotated speech by Russian heritage speakers in the U.S. and monolingually raised Russian speakers. The corpus contains data elicited in formal and informal communicative situations, by male/female and adolescent/adult speakers. The resource is presented with its architecture and annotation, and it is shown how it is used for the analysis of intonational features of spontaneous mono- and bilingual Russian speech. The analyses investigate the length of intonation phrases, types and number of pitch accents, and boundary tones. It emerges that the speaker groups do not differ in the inventory of pitch accents and boundary tones or in the relative frequency of these tonal events. However, they do differ in the length of intonation phrases (IPs), with heritage speakers showing shorter IPs also in the informal communicative situation. Both groups also differ concerning the number of pitch accents used on content words, with heritage speakers using more pitch accents than monolingually raised speakers. The results are discussed with respect to register differentiation and differences in prosodic density across both speaker groups.

show abstract

Section: Intonation Phrasessupporting

confidence: 91%

Section: Corpus Studies On Russian Intonationmentioning

confidence: 99%

See 1 more Smart Citation

Intonational Features of Spontaneous Narrations in Monolingual and Heritage Russian in the U.S.—An Exploration of the RUEG Corpus

Zerbian,

Zuban,

Klotz

2023

Languages

View full text Add to dashboard Cite

show abstract

“…Despite the fact that final accent is the most salient intonational cue, prenuclear intonation plays the role in distinguishing statements and questions in German [4], Greek [5] and Russian [6]. There is published evidence for European Portuguese [7] and Russian [8] that post-nuclear intonation can coincide with nuclear intonation.…”

Section: Introductionmentioning

confidence: 99%

Distributed representation of melodic contours

Kocharov¹,

Menshikova²

2018

Speech Prosody 2018

View full text Add to dashboard Cite

We introduce a new computational model for melodic contours-melody embeddings. It is based on the approach of distributional semantics where embeddings represent units as continuous vectors in a multi-dimensional space based on hypothesis that units with similar meaning are used in similar contexts. This paradigm is applied to melodic contours and their segments. Melodic contours are represented by vectors of the same dimensionality independent on their length and shape. We successfully evaluated the ability of the proposed model to measure the distance between melodic contours. The results of applying the model for a task of prominent words detection have not showed the improvement over traditional prosodic features. Nevertheless we assume the model to be very promising. The possible applications for the proposed unsupervised prosodic model include processing of speech of underresourced languages, modelling prosodic variability for textto-speech synthesis, recognition and classification of prosodic events by means of deep-learning algorithms.

show abstract

“…The prosody of Czech language is described in [9] and [10]. Other information on prosody of Finnish and Russian may be found in [11], [12]. The prosodic diversity of these languages makes them a good choice for our experiments.…”

Section: Introductionmentioning

confidence: 99%

Language-Dependent Melody Embeddings

Kocharov

Menshikova

2018

Interspeech 2018

View full text Add to dashboard Cite

The paper explores the perspectives of applying the distributional approach to prosodic typology of languages. The method discussed here is an adaptation of the distributional semantics approach, as suggested by Mikolov, to melodic features of speech. The paper contains a detailed description of the new method, as well as a comparison of five European languages (English, Czech, German, Russian, and Finnish) in terms of melody embeddings. The total amount of speech data was over 500 hours. The experimental results show that melody embeddings are language dependent. The proposed melody embedding model has shown reasonable results in language comparison.

show abstract

Prosodic annotation in the new corpus of Russian spontaneous speech CoRuSS

Cited by 11 publications

References 0 publications

Intonational Features of Spontaneous Narrations in Monolingual and Heritage Russian in the U.S.—An Exploration of the RUEG Corpus

Intonational Features of Spontaneous Narrations in Monolingual and Heritage Russian in the U.S.—An Exploration of the RUEG Corpus

Distributed representation of melodic contours

Language-Dependent Melody Embeddings

Contact Info

Product

Resources

About