Patents by Inventor Ivaylo Andreanov Enchev

Ivaylo Andreanov Enchev has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • On-device custom wake word detection
    Patent number: 11798535
    Abstract: Generally discussed herein are devices, systems, and methods for on-device detection of a wake word. A device can include a memory including model parameters that define a custom wake word detection model, the wake word detection model including a recurrent neural network transducer (RNNT) and a lookup table (LUT), the LUT indicating a hidden vector to be provided in response to a phoneme of a user-specified wake word, a microphone to capture audio, and processing circuitry to receive the audio from the microphone, determine, using the wake word detection model, whether the audio includes an utterance of the user-specified wake word, and wake up a personal assistant after determining the audio includes the utterance of the user-specified wake word.
    Type: Grant
    Filed: September 14, 2021
    Date of Patent: October 24, 2023
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Emilian Stoimenov, Rui Zhao, Kaustubh Prakash Kalgaonkar, Ivaylo Andreanov Enchev, Khuram Shahid, Anthony Phillip Stark, Guoli Ye, Mahadevan Srinivasan, Yifan Gong, Hosam Adel Khalil
  • ON-DEVICE CUSTOM WAKE WORD DETECTION
    Publication number: 20210407498
    Abstract: Generally discussed herein are devices, systems, and methods for on-device detection of a wake word. A device can include a memory including model parameters that define a custom wake word detection model, the wake word detection model including a recurrent neural network transducer (RNNT) and a lookup table (LUT), the LUT indicating a hidden vector to be provided in response to a phoneme of a user-specified wake word, a microphone to capture audio, and processing circuitry to receive the audio from the microphone, determine, using the wake word detection model, whether the audio includes an utterance of the user-specified wake word, and wake up a personal assistant after determining the audio includes the utterance of the user-specified wake word.
    Type: Application
    Filed: September 14, 2021
    Publication date: December 30, 2021
    Inventors: Emilian Stoimenov, Rui Zhao, Kaustubh Prakash Kalgaonkar, Ivaylo Andreanov Enchev, Khuram Shahid, Anthony Phillip Stark, Guoli Ye, Mahadevan Srinivasan, Yifan Gong, Hosam Adel Khalil
  • On-device custom wake word detection
    Patent number: 11132992
    Abstract: Generally discussed herein are devices, systems, and methods for on-device detection of a wake word. A device can include a memory including model parameters that define a custom wake word detection model, the wake word detection model including a recurrent neural network transducer (RNNT) and a lookup table (LUT), the LUT indicating a hidden vector to be provided in response to a phoneme of a user-specified wake word, a microphone to capture audio, and processing circuitry to receive the audio from the microphone, determine, using the wake word detection model, whether the audio includes an utterance of the user-specified wake word, and wake up a personal assistant after determining the audio includes the utterance of the user-specified wake word.
    Type: Grant
    Filed: July 25, 2019
    Date of Patent: September 28, 2021
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Emilian Stoimenov, Rui Zhao, Kaustubh Prakash Kalgaonkar, Ivaylo Andreanov Enchev, Khuram Shahid, Anthony Phillip Stark, Guoli Ye, Mahadevan Srinivasan, Yifan Gong, Hosam Adel Khalil
  • ON-DEVICE CUSTOM WAKE WORD DETECTION
    Publication number: 20200349927
    Abstract: Generally discussed herein are devices, systems, and methods for on-device detection of a wake word. A device can include a memory including model parameters that define a custom wake word detection model, the wake word detection model including a recurrent neural network transducer (RNNT) and a lookup table (LUT), the LUT indicating a hidden vector to be provided in response to a phoneme of a user-specified wake word, a microphone to capture audio, and processing circuitry to receive the audio from the microphone, determine, using the wake word detection model, whether the audio includes an utterance of the user-specified wake word, and wake up a personal assistant after determining the audio includes the utterance of the user-specified wake word.
    Type: Application
    Filed: July 25, 2019
    Publication date: November 5, 2020
    Inventors: Emilian Stoimenov, Rui Zhao, Kaustubh Prakash Kalgaonkar, Ivaylo Andreanov Enchev, Khuram Shahid, Anthony Phillip Stark, Guoli Ye, Mahadevan Srinivasan, Yifan Gong, Hosam Adel Khalil