Name
Abstract | ||
|---|---|---|
In this paper, we propose a novel end-to-end sequence-to-sequence spoken language understanding model using an attention mechanism. It reliably selects contextual acoustic features in order to hypothesize semantic contents. An initial architecture capable of extracting all pronounced words and concepts from acoustic spans is designed and tested. With a shallow fusion language model, this system reaches a 13.6 concept error rate (CER) and an 18.5 concept value error rate (CVER) on the French MEDIA corpus, achieving an absolute 2.8 points reduction compared to the state-of-the-art. Then, an original model is proposed for hypothesizing concepts and their values. This transduction reaches a 15.4 CER and a 21.6 CVER without any new type of context. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1109/ICASSP39728.2021.9413581 | 2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021) |
Keywords | DocType | Citations |
spoken language understanding, neural networks, attention mechanisms, sequence-to-sequence, transfer learning | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 7 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Valentin Pelloin | 1 | 0 | 0.34 |
| Nathalie Camelin | 2 | 39 | 14.29 |
| Antoine Laurent | 3 | 43 | 12.04 |
| Renato De Mori | 4 | 960 | 161.75 |
| Antoine Caubrière | 5 | 3 | 3.48 |
| Yannick Estève | 6 | 298 | 50.89 |
| Sylvain Meignier | 7 | 650 | 49.58 |