Cerence. has been granted a patent for a method that enables user devices to recognize wake-up words in a target language. The process involves analyzing acoustic inputs, comparing speech unit sequences, and adapting the acoustic model based on user-specific data, enhancing language recognition capabilities. GlobalData’s report on Cerence gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
According to GlobalData’s company profile on Cerence, Smart voice assistant was a key innovation area identified from patents. Cerence's grant share as of June 2024 was 71%. Grant share is based on the ratio of number of grants to total number of patents.
Method for recognizing wake-up words in different languages
The granted patent US12039980B2 outlines a method and user device designed to enhance speech recognition capabilities, particularly for wake-up words in a target language. The method involves receiving an initial acoustic input from a user speaking a wake-up word, which is then processed to derive acoustic features. These features are compared against a sequence of speech units generated from a second acoustic input of the same wake-up word. The acoustic model utilized for this process is trained on data from a source language that differs from the target language. The method also includes adapting the acoustic model based on the comparison results, allowing for improved recognition accuracy tailored to the user's speech patterns.
Additionally, the patent details various enhancements to the method, such as storing the first sequence of speech units as a reference, adapting the model to account for background noise, and performing speaker adaptation using techniques like maximum a posteriori (MAP) adaptation and maximum likelihood linear regression (MLLR). The user device is equipped with microphones and processors that facilitate these functions, enabling it to recognize commands while in a low power mode without needing to exit that state. The claims also cover the ability to handle different types of noise in the acoustic inputs and the storage of enrollment sequences to further refine the recognition process. Overall, the patent presents a comprehensive approach to improving user-device interaction through advanced speech recognition techniques.
To know more about GlobalData’s detailed insights on Cerence, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
