Mapping between speech signal and transcript

A computer program product for mapping between a speech signal and a transcript of the speech signal, the computer program product comprising a non-transitory computer readable storage medium having program code embodied therewith, the program code executable to:

• obtain the speech signal and the transcript of the speech signal;
• segment the speech signal to obtain multiple sets of utterance candidates for the speech signal, wherein respective ones of the multiple sets of utterance candidates have different utterance candidates;
• segment the transcript of the speech signal to one set of utterance-like units for the transcript;
• perform automatic speech recognition of the multiple sets of the utterance candidates to obtain multiple sets of recognized texts for the speech signal;
• convert the multiple sets of the recognized texts into multiple sets of estimated phone sequences;
• convert the one set of utterance-like units into one set of reference phone sequences;
• calculate costs of correspondences between respective ones of the multiple sets of the estimated phone sequences and the one set of the reference phone sequences;
• determine a smallest cost set of the estimated phone sequences from the multiple sets of estimated phone sequences;
• select a partial series of the estimated phone sequences, from the smallest cost set of the estimated phone sequences, by excluding one or more estimated phone sequences with high local mapping costs from the smallest cost set of the estimated phone sequences;
• and
• mapping data which includes the partial series of the estimated phone sequences and a corresponding series in the one set of the reference phone sequences, wherein the corresponding series corresponds to the partial series of the estimated phone sequences.

A computer system for mapping between a speech signal and a transcript of the speech signal, the computer system comprising:

• one or more processors, one or more computer readable tangible storage devices, and program instructions stored on at least one of the one or more computer readable tangible storage devices for execution by at least one of the one or more processors, the program instructions executable to:
• obtain the speech signal and the transcript of the speech signal;
• segment the speech signal to obtain multiple sets of utterance candidates for the speech signal, wherein respective ones of the multiple sets of utterance candidates have different utterance candidates;
• segment the transcript of the speech signal to one set of utterance-like units for the transcript;
• perform automatic speech recognition of the multiple sets of the utterance candidates one or more segmented speech signals to obtain multiple sets of recognized texts for the speech signal;
• convert the multiple sets of the recognized texts into multiple sets of estimated phone sequences;
• convert the one set of utterance-like units into one set of reference phone sequences;
• calculate costs of correspondences between respective ones of the multiple sets of the estimated phone sequences and the one set of the reference phone sequences;
• determine a smallest cost set of the estimated phone sequences from the multiple sets of estimated phone sequences;
• select a partial series of the estimated phone sequences, from the smallest cost set of the estimated phone sequences, by excluding one or more estimated phone sequences with high local mapping costs from the smallest cost set of the estimated phone sequences;
• and
• mapping data which includes the partial series of the estimated phone sequences and a corresponding series in the one set of the reference phone sequences, wherein the corresponding series corresponds to the partial series of the estimated phone sequences.