Name
Title | ||
|---|---|---|
Confidence-Features And Confidence-Scores For Asr Applications In Arbitration And Dnn Speaker Adaptation |
Abstract | ||
|---|---|---|
Speech recognition confidence-scores quantitatively represent correctness of decoded utterances in a [0,1] range. Confidences have primarily been used to filter out recognitions with scores below a threshold. They have also been used in other speech applications in e.g. arbitration, ROVER, and high-quality data selection for model training etc. Confidence-scores are computed from a rich set of confidence-features in the speech recognition engine. While many speech applications consume confidence scores, we haven't seen adequate focus on directly consuming confidence-features in applications. In this work we build a thesis that additionally consuming confidence-features can provide big gains across confidence-related tasks. We demonstrate this for arbitration application, where we obtain 31% relative reduction in arbitration metric. We additionally demonstrate a novel application of confidence-scores in deep-neural-network (DNN) adaptation, where we strongly improve the relative reduction in word-error-rate (WER) for speaker adaptation on limited data. |
Year | Venue | Keywords |
|---|---|---|
2015 | 16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5 | Speech recognition, Confidence scores, Confidence predictors, Classifier, MLP |
Field | DocType | Citations |
Computer science,Speech recognition,Arbitration,Speaker adaptation | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 6 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Kshitiz Kumar | 1 | 95 | 10.82 |
| Ziad Al Bawab | 2 | 25 | 2.93 |
| Yong Zhao | 3 | 127 | 13.62 |
| Chaojun Liu | 4 | 1 | 1.02 |
| Benoît Dumoulin | 5 | 0 | 0.68 |
| Yifan Gong | 6 | 1332 | 135.58 |