Cerence Operating Company

Nuance Communications, Inc.

Jianzhong Teng

Xiao-Lin Ren

Yi Gao

Yaxin Zhang

Xingui Zeng

Xiaolin Ren

A method for a user device, including receiving a first acoustic input of a user speaking a wake-up word in the target language; providing a first acoustic feature derived from the first acoustic input to an acoustic model stored on the user device to obtain a first sequence of speech units corresponding to the wake-up word spoken by the user in the target language, the acoustic model trained on a corpus of training data in a source language different than the target language; receiving a second acoustic input including the wake-up word in the target language; providing a second acoustic feature derived from the second acoustic input to the acoustic model to obtain a second sequence of speech units corresponding to the wake-up word in the target language; and comparing the first and second sequences of speech units to recognize the wake-up word in the target language.

Techniques for language independent wake-up word detection

A user device configured to perform wake-up word detection in a target language. The user device comprises at least one microphone () configured to obtain acoustic information from the environment of the user device, at least one computer readable medium () storing an acoustic model () trained on a corpus of training data () in a source language different than the target language, and storing a first sequence of speech units obtained by providing acoustic features () derived from audio comprising the user speaking a wake-up word in the target language to the acoustic model (), and at least one processor () coupled to the at least one computer readable medium () and programmed to perform receiving, from the at least one microphone (), acoustic input from the user speaking in the target language while the user device is operating in a low-power mode, applying acoustic features derived from the acoustic input to the acoustic model () to obtain a second sequence of speech units corresponding to the acoustic input, determining if the user spoke the wake-up word at least in part by comparing the first sequence of speech units to the second sequence of speech units, and exiting the low-power mode if it is determined that the user spoke the wake-up word.

Methods and an apparatus for performing feature extraction on speech in a microphone signal with embedded noise processing to reduce the amount of processing are provided. In embodiments, feature extraction and the noise estimate use an output of the same Fourier Transform, such that the noise filtering of the speech is embedded with the feature extraction of the speech.

Methods and apparatus for ASR with embedded noise reduction

Techniques disclosed herein include systems and methods of automated speech recognition (ASR) for voice destination entry (VDE) include open voice searching (natural language searching) of destinations. A first part uses a server-based automated speech recognizer. The second part is client-based automatic speech recognition (ASR) processing. Thus, techniques include a hybrid VDE solution that provides users with an accurate and flexible way to use speech recognition technologies. A server-based speech recognizer executes the open-search task, while a client-based recognizer refines the results from the server to deliver an optimized result. This system and method significantly improves recognition accuracy for dictation engine based POI search of Chinese Mandarin input and input from other languages. Moreover, the methods herein largely improve the user experience by allowing users to say a partial POI name, and abbreviation, or even say a POI name in a reversed word order.

Growing community of inventors

Shanghai, China

Jianzhong Teng