It is estimated that 300 million people use Google every day in their quest to find information on the web. Imagine having the ability to not only search online text files for specific words but also to search audio files for specific sounds. Thanks in part to Messiah College Engineering Professor Randy Fish, that technology is now patented.
Fish and a team of five others have spent the past seven years perfecting and patenting “audio hot spotting,” a search engine technology that allows users to search audio files for a specific word or phrase, a particular speaker, a speaker’s tone, or characteristics like laughter and applause.
This type of “audio hot spotting” might, for example, be helpful to a user who only wants to hear the portions of a speech that caused an audience to react with applause or laughter.
The challenge, according to Fish, was developing the artificial intelligence algorithms able to discern emotion in a voice, regardless of whether the voice is male or female, and to recognize applause and laughter generically.
Other systems attempt this type of audio searching, Fish says, but only this patented system gives users access to emotions and non-speaking sounds in addition to searching on what was said or who said it.