摘要 |
This invention relates to a method of mining spoken audio data for one or more search terms comprising performing a phonetic search of the audio data to identify likely matches to the search term(s) and producing textual data corresponding to portions of the spoken audio data including a likely match. A phonetic index of data corresponding to the spoken audio data may be created before the phonetic search. The selected likely matching portions may be processed using a large vocabulary speech recogniser (LVCSR). The large vocabulary speech recogniser may derive textual data which can be used for further processing or may be presented to a user. The present invention therefore combines the benefit of phonetic searching of audio data with the advantages of large vocabulary speech recognition. |