Surety is performing system maintenance this weekend. Electronic date stamps on new Prior Art Database disclosures may be delayed.
Browse Prior Art Database

Personalised announcement detection for realtime playback in headphones

IP.com Disclosure Number: IPCOM000244581D
Publication Date: 2015-Dec-23
Document File: 1 page(s) / 33K

Publishing Venue

The IP.com Prior Art Database


This article describes a way in which a user wearing headphones can be alerted of relevant notices suggested by their day by day actions.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 66% of the total text.

Page 01 of 1

Personalised announcement detection for realtime playback in headphones

More people are becoming immersed in technology in an increasing number of situations. A common form of this is listening to music or watching a movie on your smart device using headphones. When wearing headphones you can easily miss announcements and other auditory information relevant to your day, e.g. announcements at train stations, airports, on public transport, traffic crossings etc.

    This system will allow the user to continue to use headphones and listen to audio as normal, detect any information relevant to them, pause the soundtrack the user is listening to and then relay the information to them.

The steps for how this method would work are detailed below:

Step 1. Microphones are embedded into the exterior of the headphones
Step 2. The headphones continuously record audio, storing only snippets of a short specific length (e.g. announcements are generally shorter than a minute long) in a continuous way by overwriting previous data. This ensures that storage requirements are small.

Step 3. Use a filter to remove background noise from the audio.

    Step 4. Look for auditory matches against either: a. Known signals such as traffic crossing sounds, fire alarms, emergency vehicle sirens etc.

    b. Keywords extracted from your internet history and purchases, for example flight details, hotel reservations

         --> use entity extraction from the internet history details and the auditory input. Use speec...