On-line captioning of TV-programs for the hearing impaired

Harborg, Erik; Holter, Trym; Johnsen, Magne Hallstein; Svendsen, Torbjørn

doi:10.21437/eurospeech.1999-146

Cited by 1 publication

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The OOV rate for these experiments is 2.2%. In contrast, our previously reported results [3] was obtained using a closed vocabulary of 12K words, and a bigram extracted from the 120K words annotated speech database, which implies that the test data was applied in the language model design. However, the test set is exactly the same.…”

Section: Experiments and Resultsmentioning

confidence: 81%

“…In our previous experiments [3], we had available a recorded speech database containing a total of about 17 hours of speech, uttered by 26 different speakers. We have now completed the recordings of the speech database with 30 additional speakers.…”

Section: Databases For Text and Speechmentioning

confidence: 99%

“…In our previous experiments [3] we designed various N-gram language models based on the annotated speech databases within a closed vocabulary framework. In this paper, we address the more realistic situation with an open vocabulary.…”

Section: Language Modellingmentioning

confidence: 99%

“…In this paper we describe our current efforts in developing a system for generation of closed captions for broadcast of live TVprograms, based on automatic speech recognition (ASR) [1] [2]. Here, we extend the work described in [3]. This application is aimed to be a new service from the Norwegian Broadcasting Corporation (NRK), offering the hearing impaired a possibility to enjoy live broadcast TV-programs.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

ASR-based subtitling of live TV-programs for the hearing impaired

Holter¹,

Harborg²,

Johnsen³

et al. 2000

6th International Conference on Spoken Language Processing (ICSLP 2000)

Self Cite

View full text Add to dashboard Cite

A system for on-line generation of closed captions (subtitles) for broadcast of live TV-programs is described. During broadcast, a commentator formulates a possibly condensed, but semantically correct version of the original speech. These compressed phrases are recognized by a continuous speech recognizer, and the resulting captions are fed into the teletext system. This application will provide the hearing impaired with an option to read captions for live broadcast programs, i.e., when off-line captioning is not feasible.The main advantage in using a speech recognizer rather than a stenography-based system (e.g., Velotype) is the relaxed requirements for commentator training. Also, the amount of text generated by a system based on stenography tends to be large, thus making it harder to read.

show abstract

Section: Experiments and Resultsmentioning

confidence: 81%

Section: Databases For Text and Speechmentioning

confidence: 99%

Section: Language Modellingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations