Back to results
US 2016/0080295 A1Published application
System and Method for Voice Networking
OwnerStephen Daviesinventor
Life of this patent
Estimating the term…
Loading patent…
Loading patent…
OwnerStephen Daviesinventor
Estimating the term…
A system and method for voice based social networking is disclosed. The system receives a voice message (and frequently an image) and ultimately delivers it to one or multiple users, placing it within an ongoing context of conversations. The voice and image may be recorded by various devices and the data transmitted in a variety of formats. An alternative implementation places some system functionality in a mobile device such as a smartphone or wearable device, with the remaining functionality resident in system servers attached to the internet. The system can apply rules to select and limit the voice data flowing to each user; rules prioritize the messages using context information such as user interest and user state. An image is fused to the voice message to form a comment. Additional image or voice annotation (or both) identifying the sender may be attached to the comment. Fused image(s) and voice annotation allow the user to quickly deduce the context of the comment. The system provides the means to deliver a seamless flow of a conversation without requiring user interaction to hear individual comments. Multiple conversations or parts of conversations (or both) are blended into the users acoustic channel without requiring user interaction at the level of individual comments. Advertising may also be integrated. The resulting information flow is translated into formats acceptable to the user's receiving device.
A method for voice-based social networking wherein for a temporal sequence of voice message segments, voice message segments are augmented with additional data comprising a direct indicator of the speaker for each segment, wherein said direct indicator of the speaker is one of an image representing said speaker and a segment of voice representing said speaker, and an indicator of the conversation to which each segment belongs, wherein said indicator of the conversation is one of the most recently uploaded image in the temporal sequence of voice messages of said conversation and the first image uploaded in said conversation, so that the user receiving said message segments can easily infer the context of each voice segment and follow multiple conversations thus permitting said user effective use of said network while allocating significant attention to activities outside of interacting with said network.
A method for voice-based social networking wherein, for certain periods each of which forming a time interval, voice messages and other associated data are selected for presentation to the recipient using a scoring algorithm for each message wherein said scoring algorithm incorporates a user dependent weighting of message components to produce a score that reflects the relative importance of said message, said messages are ranked by this score and the highest ranked messages are selected for forwarding to the user in said interval such that the rate of presentation of these selected messages is maintained at a rate below the smallest of the rate which the acoustic channel can support and the rate at which the recipient is judged capable of effective comprehension as determined by an assessment of the state of the user.
Means for voice-based social networking comprising a processor, sensors, an electronic network, a sensor interface, an audio store or cache, an image store or cache, a comment descriptor database, an advertising database, a selection rules engine, a comment delivery module, a display and an acoustic output device wherein:
Dependencies are inferred from the claim text.