Name
Abstract | ||
|---|---|---|
Videolectures are currently being digitised all over the world for its enormous value as reference resource. Many of these lectures are accompanied with slides. The slides offer a great opportunity for improving ASR systems performance. We propose a simple yet powerful extension to the linear interpolation of language models for adapting language models with slide information. Two types of slides are considered, correct slides, and slides automatic extracted from the videos with OCR. Furthermore, we compare both time aligned and unaligned slides. Results report an improvement of up to 3.8 % absolute WER points when using correct slides. Surprisingly, when using automatic slides obtained with poor OCR quality, the ASR system still improves up to 2.2 absolute WER points. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/ICASSP.2013.6639314 | Acoustics, Speech and Signal Processing |
Keywords | Field | DocType |
interpolation,multimedia systems,optical character recognition,video signal processing,ASR systems performance,OCR quality,absolute WER points,automatic extracted slides,correct slides,language model adaptation,language models,linear interpolation,reference resource,slide information,time aligned slides,unaligned slides,video lectures transcription,language model adaptation,video lectures | Computer science,Interpolation,Optical character recognition,Speech recognition,Linear interpolation,Language model | Conference |
ISSN | Citations | PageRank |
1520-6149 | 3 | 0.41 |
References | Authors | |
0 | 7 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Adria A. Martinez-Villaronga | 1 | 8 | 1.86 |
| Miguel A. del Agua | 2 | 8 | 1.54 |
| Jesús Andrés-Ferrer | 3 | 73 | 7.52 |
| alfons juan | 4 | 572 | 61.45 |
| Martinez-Villaronga, A. | 5 | 3 | 0.41 |
| del Agua, M.A. | 6 | 3 | 0.41 |
| Andres-Ferrer, J. | 7 | 3 | 0.41 |