The NNI Query-by-Example System for MediaEval 2014
In this paper we describe the system proposed by NNI (NWPU-NTU-I2R) team for the QUESST task within the Mediaeval 2014 evaluation. To solve the problem, we used both dynamic time warping (DTW) and symbolic search (SS) based approaches. The DTW system performs template matching using subsequence DTW algorithm and posterior representations. The symbolic search is performed on phone sequences generated by phone recognizers. For both symbolic and DTW search, partial sequence matching is performed to reduce missing rate, especially for query type 2 and 3. After fusing 9 DTW systems, 7 symbolic systems, and query length side information, we obtained 0.6023 actual normalized cross entropy (actCnxe) for all queries combined. For type 3 complex queries, we achieved 0.7252 actCnxe.
PDF
Datasets
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
|---|---|---|---|---|---|---|
| Keyword Spotting | QUESST | NNI non-filtered(for the development set) | Cnxe | 6.0905 | # 63 | |
| MinCnxe | 0.9571 | # 60 | ||||
| ATWV | 0.0768 | # 20 | ||||
| MTWV | 0.0767 | # 20 | ||||
| Keyword Spotting | QUESST | NNI Choi(for the development set) | Cnxe | 5.8940 | # 62 | |
| MinCnxe | 0.9595 | # 61 | ||||
| ATWV | 0.0692 | # 21 | ||||
| MTWV | 0.0692 | # 21 | ||||
| Keyword Spotting | QUESST | NNI Symbolic(All Queries) | Cnxe | 0.7322 | # 8 | |
| MinCnxe | 0.7293 | # 13 | ||||
| ATWV | 0.2696 | # 10 | ||||
| MTWV | 0.2717 | # 11 | ||||
| Keyword Spotting | QUESST | NNI DTW(All Queries) | Cnxe | 0.6925 | # 6 | |
| MinCnxe | 0.6816 | # 12 | ||||
| ATWV | 0.2918 | # 7 | ||||
| MTWV | 0.2974 | # 8 |
