Browse Prior Art Database

System and method of smart file searching criteria based on audio Disclosure Number: IPCOM000245776D
Publication Date: 2016-Apr-07
Document File: 5 page(s) / 101K

Publishing Venue

The Prior Art Database


In this disclosure, a method is proposed to help user add new keywords for the presentation files through analyzing the related audio of the presentation according to speaker tone and speech rate, and sum up these keywords for further search. It will help user save time and effort for the later presentation file search.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 5

System and method of smart file searching criteria based on audio


In our daily management of presentation files, we may need to add tags for the files so that we can retrieve them quickly through tags. What we do now is to manually add tags or use some tools to add tags automatically . However, these tools can only choose some keywords that already in the content of the file as tags. Usually keywords are not included in the content of files. We can only sum up the required keywords by adding them manually, but it takes much more effort.

What we can address here is:

Whether there is a smart method which can help us add tags for presentation files when the required keywords are not in the content of the files. This method can add tags for the files through analyzing the related audio of the presentation to sum up keywords we need. It helps user save time and effort.


Penny is trying to search on presentation files in her laptop, with keyword "Audio Analysis", which is the best she can think of. While this keyword text may not appear as content in the presentation files , in which they only include contents of other synonym or with strong correlation, so she may not get all the target charts by this pure text search.

However, Jack had once delivered a session with one presentation file, and an audio was recorded of the whole session. During the session, he had mentioned lots of background knowledge, including the desired keyword. We will leverage the recorded audio file, and extract potential key words, and insert them to the presentation file as tags. In this way, the presentation file will be marked as the new key word, and further searches can get this presentation file target content.

When just finished this session, Penny has good understanding on the contents of session very clearly. But after one month, she

just can remember some keywords of this session in the mind, but they may not be included in the presentation file(Penny has gotten the presentation file and the related audio of this session from Jack ), so the result is not expected by Penny. In our disclosure, we will extract new keywords from the audio according to speaker tone and speech rate, then deal with them with the existing keywords in presentation file with some modules, then add the integrated keywords into as presentation file tags for further


Page 02 of 5


a) Extend user's search scope.

b) Make the search action more accurate, more smart.

c) After more version of audio files (generated by sessions delivered by different people) provided for a single presentation file, more tags will be added, this will further assist on the search accuracy.

d) We can even provide the exact slide of the presentation file to match the search key word .