| 摘要: |
| 提出了一种基于联合概率似然得分和概率似然比得分进行语音识别确认的方法,计算搜索路径得分过程中同时考虑概率似然比得分,在给出系统最终识别结果的同时给出置信度水平.实验结果表明,该方法在大大降低识别系统误警率的同时,基本保持识别正确率不变. |
| 关键词: 似然比检验 备择模型 语音确认 语音识别 |
| DOI： |
| 分类号: |
| 基金项目:Supported by the National Natural Science Foundation of China under Grant No.60272019 (国家自然科学基金); the National Grand Fundamental Research 973 Program of China under Grant No.G199803050703 (国家重点基础研究发展规划(973)) |
|
| Continuous Speech Recognition and Verification Based on a Combination Score |
|
Yan Bin-Feng,Zhu Xiao-Yan
|
| Abstract: |
| In this paper, a speech recognition and verification method is presented based on the integration of likelihood and likelihood ratio. Speech recognition and verification is unified in one-phase framework. In the process of decoding, likelihood ratio is combined with likelihood to get the combination score for searching the final results. Experimental results show that the false-alarm rate is significantly reduced, with only a slight loss in the accuracy rate. |
| Key words: likelihood ratio test alternative model utterance verification speech recognition |
