Automatic punctuation annotation in czech broadcast news speech
Digital Library (University of West Bohemia)Published 20 September 2004Open access
Jáchym Kolář, Ján Švec, Josef Psutka
Citations32
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
This paper reports the initial experiments with automatic punctuation annotation from speech on Czech broadcast news speech using prosodic model and language model, and the model combining language model with the decision tree yielded superior results.
Abstract
Tento článek se zabývá našimi počátečními experimenty s automatickou anotací interpunkce v mluvené češtině. Použili jsme 2 statistické modely - prozodický a jazykový. Byly otestovány 2 implementace prozodického modelu - CART a MLP. Pro jazykové modelováni byl použit N-gramový model se skrytými událostmi. Kombinovaný model dosáhl na referenčních přepisech přesnosti 95.2% a F-measure 78.2%.
Keywords
Computer Science
BiometricsClassification and Regression Trees.
23,841 Citations1984Alexander Gordon, Leo Breiman +3 more
Speech CommunicationProsody-based automatic segmentation of speech into sentences and topics
420 Citations2000Elizabeth Shriberg, Andreas Stolcke +2 more
This work combines prosodic cues with word-based approaches, and evaluates performance on two speech corpora, Broadcast News and Switchboard, finding that the prosodic model achieves comparable performance with significantly less training data, and requires no hand-labeling of prosodic events.
Modeling dynamic prosodic variation for speaker verification
141 Citations1998Kemal Sönmez, Elizabeth Shriberg +2 more
This work model the speaker’s f0 movements by fitting a piecewise linear model to the f0 track to obtain a stylized f0 contour, and improves the verification performance of a cepstrum-based Gaussian mixture model system by 10%.
Automatic detection of sentence boundaries and disfluencies based on recognized words
138 Citations1998Andreas Stolcke, Elizabeth Shriberg +6 more
The problem of detecting linguistic events at interword boundaries, such as sentence boundaries and disfluency locations, in speech transcribed by an automatic recognizer is studied, based on a combination of prosodic cues modeled by decision trees, and word-based event N-gram language models.
Tagging Inflective Languages: Prediction of Morphological Categories for a Rich, Structured Tagset
124 Citations1998Jan Hajič, Barbora Hladká
Edinburgh Research Archive (University of Edinburgh)Punctuation annotation using statistical prosody models.
121 Citations2001Heidi Christensen, Yoshihiko Gotoh +1 more
A statistical finite state model that combines prosodic, linguistic and punctuation class features to generate linguistic meta-data for spoken language is presented.
A lognormal tied mixture model of pitch for prosody based speaker recognition
85 Citations1997Kemal Sönmez, Larry Heck +2 more
A statistical model of pitch is developed that allows unbiased estimation of pitch statistics from pitch tracks which are subject to doubling and/or halving and which argues by a simple correlation model and empirically demonstrate that “clean” pitch is distributed with a lognormal distribution rather than the often assumed normal distribution.
Direct modeling of prosody: an overview of applications in automatic speech processing
62 Citations2004Elizabeth Shriberg, Andreas Stolcke
This work describes a “direct modeling” approach to using prosody in various speech technology tasks that does not involve any hand-labeling or modeling of prosodic events such as pitch accents or boundary tones, and focuses on spontaneous speech from a variety of contexts.
Speech CommunicationA combined punctuation generation and speech recognition system and its performance enhancement using prosody
37 Citations2003Ji‐Hwan Kim, Philip C. Woodland
A punctuation generation system which combines prosodic information with acoustic and language model information and gives a small reduction in word error rate on the 1-best speech recognition output including punctuation is presented.
The COCOON platform (University of Paris)Tagging Inlective Languages: Prediction of Morphological Categories for a Rich Structured Tagset
33 CitationsJan Hajič, Barbora Hladká
Lecture notes in computer scienceFast and Robust Features for Prosodic Classification?
11 Citations1999Jan Buckow, Volker Warnke +4 more
With the new set of prosodic features better classification results can be achieved, the features extraction can be sped up by 64 %, and the memory requirements are even reduced by 92%.
The czech speech and prosody database both for ASR and TTS purposes
7 Citations2003Jáchym Kolář, Jan Romportl +1 more
A preparation of the first large Czech prosodic database which should be useful both in automatic speech recognition (ASR) and text-to-speech (TTS) synthesis and for building a prosodic module for the Czech high-quality synthesis.