18261449. REDUCED-LATENCY SPEECH PROCESSING simplified abstract (QUALCOMM Incorporated)

From WikiPatents
Jump to navigation Jump to search

REDUCED-LATENCY SPEECH PROCESSING

Organization Name

QUALCOMM Incorporated

Inventor(s)

Xiaoxia Dong of Shenzhen (CN)

Leonid Sheynblat of Hillsborough CA (US)

Zhijia Sun of Shenzhen (CN)

Xin Chen of Shenzhen (CN)

Sabrina Chung Man Wong of San Diego CA (US)

REDUCED-LATENCY SPEECH PROCESSING - A simplified explanation of the abstract

This abstract first appeared for US patent application 18261449 titled 'REDUCED-LATENCY SPEECH PROCESSING

Simplified Explanation

The device described in the patent application includes a first stage keyword detector and a processor. The processor is designed to initiate a change of state to enable a voice service that is in a low-power mode to be scheduled for execution on reserved processor cores, based on a signal indicating keyword detection by the first stage keyword detector before confirmation of the detection.

  • The device includes a first stage keyword detector and a processor.
  • The processor can initiate a change of state to enable a voice service in low-power mode to be scheduled for execution on reserved processor cores.
  • This change of state is based on a signal indicating keyword detection by the first stage keyword detector before confirmation of the detection.
      1. Potential Applications
  • This technology could be used in smart home devices to improve voice recognition capabilities.
  • It could also be applied in automotive systems for hands-free voice commands.
      1. Problems Solved
  • Enables efficient use of processor resources by scheduling voice services on reserved cores.
  • Improves responsiveness of voice services by initiating a change of state based on early keyword detection.
      1. Benefits
  • Enhances user experience by enabling faster execution of voice commands.
  • Optimizes power consumption by utilizing low-power mode for voice services when not in use.


Original Abstract Submitted

A device includes a first stage keyword detector and a processor. The processor is configured to, based on a signal indicating a keyword detection by the first stage keyword detector and prior to confirmation of the keyword detection, initiate a change of state to enable a voice service that is in a low-power mode to be scheduled for execution on one or more reserved processor cores.

Retrieved from "http://wikipatents.org/index.php?title=18261449._REDUCED-LATENCY_SPEECH_PROCESSING_simplified_abstract_(QUALCOMM_Incorporated)&oldid=36705"