Transcription System for Semi-Spontaneous Estonian Speech

被引：2

作者：

Alumaee, Tanel ^{[1
]}

机构：

[1] Tallinn Univ Technol, Inst Cybernet, EE-19086 Tallinn, Estonia

来源：

HUMAN LANGUAGE TECHNOLOGIES: THE BALTIC PERSPECTIVE | 2012年 / 247卷

关键词：

Estonian; speech recognition; compound words; RECOGNITION;

D O I：

10.3233/978-1-61499-133-5-10

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper describes a speech-to-text system for semi-spontaneous Estonian speech. The system is trained on about 100 hours of manually transcribed speech and a 300M word text corpus. Compound words are split before building the language model and reconstructed from recognizer output using a hidden event N-gram model. We use a three pass transcription strategy with unsupervised speaker adaptation between individual passes. The system achieves a word error rate of 34.6% on conference speeches and 25.6% on radio talk shows.

引用

页码：10 / 17

页数：8

共 50 条

[1] Prosodic Characteristics of Semi-Spontaneous Speech
Bialyk, Olena
PSYCHOLINGUISTICS, 2018, 23 (02): : 40 - 50
[2] Semi-spontaneous Appearances in Hypnosis
Alrutz, Sydney
ZEITSCHRIFT FUR PSYCHOLOGIE UND PHYSIOLOGIE DER SINNESORGANE, 1909, 52 : 425 - 460
[3] Automatic Analysis of The Prosodic Variations in Parkinsonian Read and Semi-Spontaneous Speech
De Looze, Celine
Ghio, Alain
Scherer, Stefan
Pouchoulin, Gilles
Viallet, Francois
PROCEEDINGS OF THE 6TH INTERNATIONAL CONFERENCE ON SPEECH PROSODY, VOLS I AND II, 2012, : 71 - 74
[4] DESCRIBING NON-VOLUNTARY REPETITION OF SEGMENTS IN SEMI-SPONTANEOUS SPEECH
Montes de Oca, Domingo Roman
Cofre Vergara, Valeria
Quezada Gaponov, Camilo
Matas Crespo, Jose
Lluma i Fuentes, Jordi
LOGOS-REVISTA DE LINGUISTICA FILOSOFIA Y LITERATURA, 2011, 21 (02): : 4 - 16
[5] Ecologically valid speech collection in behavioral research: The Ghent Semi-spontaneous Speech Paradigm (GSSP)
Van der Donckt, Jonas
Kappen, Mitchel
Degraeve, Vic
Demuynck, Kris
Vanderhasselt, Marie-Anne
Van Hoecke, Sofie
BEHAVIOR RESEARCH METHODS, 2024, 56 (06) : 5693 - 5708
[6] Advanced Rich Transcription System for Estonian Speech
Alumae, Tanel
Tilk, Ottokar
Asadullah
HUMAN LANGUAGE TECHNOLOGIES - THE BALTIC PERSPECTIVE, BALTIC HLT 2018, 2018, 307 : 1 - 8
[7] Empty onset repairs in the semi-spontaneous speech of Spanish child and adult heritage speakers
Repiso-Puigdelliura, Gemma
INTERNATIONAL JOURNAL OF BILINGUALISM, 2021, 25 (05) : 1311 - 1326
[8] Preparation of Bicelles Using the Semi-spontaneous Method
Watanabe, Yuki
Aramaki, Kenji
Kadomatsu, Yuya
Tanaka, Ken
Konno, Yoshikazu
CHEMISTRY LETTERS, 2016, 45 (05) : 558 - 560
[9] ANALYSIS OF SEMI-SPONTANEOUS SPEECH BEFORE, DURING AND AFTER AWAKE CRANIOTOMY: A CASE STUDY
Gommers, E. C.
Collee, K. E.
Vincent, A. J. P. E.
Bos, E. M.
Dirven, C. M. F.
Koekkoek, S. K.
Kruizinga, P.
Satoer, D. D.
NEURO-ONCOLOGY, 2022, 24
[10] Formulation of bicelles with cholesterol using a semi-spontaneous method
Kamimoto-Kuroki, Junko
Yamashita, Mika
Tanaka, Ken
Kadomatsu, Yuya
Tsukamoto, Daisuke
Aramaki, Kenji
Adachi, Keita
Konno, Yoshikazu
COLLOIDS AND SURFACES A-PHYSICOCHEMICAL AND ENGINEERING ASPECTS, 2020, 606

← 1 2 3 4 5 →