Transcribing radio news

Author

Kubala, Francis ; Anastasakos, Tasos ; Jin, Hubert ; Nguyen, Long ; Schwartz, Richard

Author_Institution

BBN Syst. & Technol. Corp., Cambridge, MA, USA

Volume

2

fYear

1996

fDate

3-6 Oct 1996

Firstpage

598

Abstract

We have recently extended the capabilities of BBN´s large-vocabulary discrete-utterance speech recognition system (BYBLOS) to operate on raw audio recordings of radio news programming. The recordings are given to the system as large monolithic waveforms without any additional side-information. Our goal is to transcribe all speech in the input with the highest accuracy possible. The problem is very challenging because radio news programming has frequent changes in speaker, speaking style, dialect, accent, topic, channel and environmental conditions. Furthermore, the monolithic input presents new problems for recognition algorithms and language models since all useful boundaries (such as speaker turns or sentence ends) are unknown

Keywords

audio recording; radio broadcasting; speech recognition; BBN Systems and Technologies; BYBLOS; accent; accuracy; channel; dialect; environmental conditions; language models; large monolithic waveforms; large-vocabulary discrete-utterance speech recognition system; monolithic input; radio news programming; radio news transcription; raw audio recordings; sentence ends; speaker turns; speaking style; topic; unknown boundaries; Audio recording; Background noise; Bandwidth; Hidden Markov models; Natural languages; Speech recognition; System testing; Telephony; Training data; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on

Conference_Location

Philadelphia, PA

Print_ISBN

0-7803-3555-4

Type

conf

DOI

10.1109/ICSLP.1996.607432

Filename

607432