专利名称：HYPERARTICULATION DETECTION IN
REPETITIVE VOICE QUERIES USING PAIRWISECOMPARISON FOR IMPROVED SPEECHRECOGNITION
发明人：Ranjitha Gurunath Kulkarni,Ahmed Moustafa
El Kholy,Ziad Al Bawab,Noha Alon,ImedZitouni
申请号：US15624451申请日：20170615
公开号：US20180254035A1公开日：20180906
专利附图：
摘要：Automatic speech recognition systems can benefit from cues in user voice suchas hyperarticulation. Traditional approaches typically attempt to define and detect anabsolute state of hyperarticulation, which is very difficult, especially on short voicequeries. This disclosure provides for an approach for hyperarticulation detection usingpair-wise comparisons and on a real-world speech recognition system. The disclosedapproach uses delta features extracted from a pair of repetitive user utterances. Theimprovements provided by the disclosed systems and methods include improvements inword error rate by using hyperarticulation information as a feature in a second pass N-best hypotheses rescoring setup.
申请人：Microsoft Technology Licensing, LLC
地址：Redmond WA US
国籍：US
更多信息请下载全文后查看
因篇幅问题不能全部显示，请点此查看更多更全内容