Samsung electronics co., ltd. (20240339123). SYSTEM AND METHOD FOR KEYWORD SPOTTING IN NOISY ENVIRONMENTS simplified abstract

SYSTEM AND METHOD FOR KEYWORD SPOTTING IN NOISY ENVIRONMENTS

Organization Name

samsung electronics co., ltd.

Inventor(s)

Chou-Chang Yang of San Jose CA (US)

Yashas Malur Saidutta of Menlo Park CA (US)

Rakshith Sharma Srinivasa of Sunnyvale CA (US)

Ching-Hua Lee of Mountain View CA (US)

Yilin Shen of San Jose CA (US)

Hongxia Jin of San Jose CA (US)

SYSTEM AND METHOD FOR KEYWORD SPOTTING IN NOISY ENVIRONMENTS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240339123 titled 'SYSTEM AND METHOD FOR KEYWORD SPOTTING IN NOISY ENVIRONMENTS

Simplified Explanation: The patent application describes a method that processes audio input to detect speech and keywords, enhancing the audio for better keyword classification.

Key Features and Innovation:

• Receiving audio input and generating a noisy time-frequency representation.
• Providing the representation to a noise management model to predict a denoising mask and signal presence probability map.
• Determining an enhanced spectrogram using the denoising mask.
• Providing the enhanced spectrogram and signal presence probability map to a keyword classification model.
• Transmitting the audio input to a downstream application if a keyword is detected.

Potential Applications: This technology can be used in speech recognition systems, keyword detection in audio content, and enhancing audio quality for better classification.

Problems Solved: The method addresses the challenges of noise interference in audio signals, improving the accuracy of speech and keyword detection in audio data.

Benefits:

• Enhanced speech and keyword detection accuracy.
• Improved audio quality for downstream applications.
• Efficient noise reduction in audio processing.

Commercial Applications: The technology can be applied in voice-controlled devices, transcription services, call center analytics, and audio content indexing for search engines.

Prior Art: Prior research in speech recognition, audio signal processing, and keyword detection can provide insights into related technologies and approaches.

Frequently Updated Research: Stay updated on advancements in speech processing, machine learning models for audio analysis, and keyword detection algorithms for continuous improvement in this field.

Questions about Audio Processing Technology: 1. How does the noise management model predict the denoising mask and signal presence probability map? 2. What are the key factors that contribute to the accuracy of keyword classification in this method?

Original Abstract Submitted

a method includes receiving an audio input and generating a noisy time-frequency representation based on the audio input. the method also includes providing the noisy time-frequency representation to a noise management model trained to predict a denoising mask and a signal presence probability (spp) map indicating a likelihood of a presence of speech. the method further includes determining an enhanced spectrogram using the denoising mask and the noisy time-frequency representation. the method also includes providing the enhanced spectrogram and the spp map as inputs to a keyword classification model trained to determine a likelihood of a keyword being present in the audio input. in addition, the method includes, responsive to determining that a keyword is in the audio input, transmitting the audio input to a downstream application associated with the keyword.