Symbolic Weighted Language Models, Quantitative Parsing and Automated Music Transcription - Archive ouverte HAL Access content directly
Conference Papers Year : 2022

Symbolic Weighted Language Models, Quantitative Parsing and Automated Music Transcription

Abstract

We study several classes of symbolic weighted formalisms: automata (swA), transducers (swT) and visibly pushdown extensions (swVPA, swVPT). They combine the respective extensions of their symbolic and weighted counterparts, allowing a quantitative evaluation of words over a large or infinite input alphabet. We present properties of closure by composition, the computation of transducer-defined distances between nested words and languages, as well as a PTIME 1-best search algorithm for swVPA. These results are applied to solve in PTIME a variant of parsing over infinite alphabets. We illustrate this approach with a motivating use case in automated music transcription.
Fichier principal
Vignette du fichier
main.pdf (499.01 Ko) Télécharger le fichier
CIAA22-FJLRdLN.pdf (6.27 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Format : Presentation
Comment : presentation slides

Dates and versions

hal-03647675 , version 1 (20-04-2022)
hal-03647675 , version 2 (07-06-2022)

Identifiers

Cite

Florent Jacquemard, Lydia Rodriguez de La Nava. Symbolic Weighted Language Models, Quantitative Parsing and Automated Music Transcription. CIAA 2022 - International Conference on Implementation and Application of Automata, Jun 2022, Rouen, France. pp.67-79, ⟨10.1007/978-3-031-07469-1_5⟩. ⟨hal-03647675v2⟩
202 View
115 Download

Altmetric

Share

Gmail Facebook X LinkedIn More