Google llc (20240096326). UTTERANCE CLASSIFIER simplified abstract
Contents
- 1 UTTERANCE CLASSIFIER
UTTERANCE CLASSIFIER
Organization Name
Inventor(s)
Nathan David Howard of Mountain View CA (US)
Gabor Simko of Santa Clara CA (US)
Maria Carolina Parada San Martin of Boulder CO (US)
Ramkarthik Kalyanasundaram of Cupertino CA (US)
Guru Prakash Arumugam of Sunnyvale CA (US)
Srinivas Vasudevan of Mountain View CA (US)
UTTERANCE CLASSIFIER - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240096326 titled 'UTTERANCE CLASSIFIER
Simplified Explanation
The method described in the abstract involves using a neural network-based utterance classifier to determine if a spoken utterance is directed towards an automated assistant server or not. If the utterance is directed towards the server, instructions are generated for the server to respond.
- Neural network-based utterance classifier with LSTM layers
- Trained on negative training examples
- Determines if spoken utterance is directed towards automated assistant server
- Generates instructions for server response
Potential Applications
This technology can be applied in:
- Voice-controlled virtual assistants
- Customer service chatbots
- Speech recognition systems
Problems Solved
This technology helps in:
- Improving accuracy of identifying user intent
- Enhancing user experience with automated systems
- Streamlining communication with virtual assistants
Benefits
The benefits of this technology include:
- Efficient handling of spoken commands
- Personalized responses from automated systems
- Increased productivity in voice-based interactions
Potential Commercial Applications
The potential commercial applications of this technology include:
- Virtual assistant devices for homes and offices
- Call center automation software
- Voice-activated smart devices
Possible Prior Art
One possible prior art for this technology could be:
- Speech recognition systems used in virtual assistants like Siri or Alexa
What are the limitations of the neural network-based utterance classifier described in the abstract?
The abstract does not mention the accuracy rate of the classifier in determining if a spoken utterance is directed towards the automated assistant server.
How does the negative training examples of spoken utterances not directed towards an automated assistant server contribute to the effectiveness of the neural network-based classifier?
The abstract does not explain how the negative training examples help in improving the classifier's ability to differentiate between utterances directed towards the automated assistant server and those that are not.
Original Abstract Submitted
a method includes receiving a spoken utterance that includes a plurality of words, and generating, using a neural network-based utterance classifier comprising a stack of multiple long-short term memory (lstm) layers, a respective textual representation for each word of the of the plurality of words of the spoken utterance. the neural network-based utterance classifier trained on negative training examples of spoken utterances not directed toward an automated assistant server. the method further including determining, using the respective textual representation generated for each word of the plurality of words of the spoken utterance, that the spoken utterance is one of directed toward the automated assistant server or not directed toward the automated assistant server, and when the spoken utterance is directed toward the automated assistant server, generating instructions that cause the automated assistant server to generate a response to the spoken utterance.
- Google llc
- Nathan David Howard of Mountain View CA (US)
- Gabor Simko of Santa Clara CA (US)
- Maria Carolina Parada San Martin of Boulder CO (US)
- Ramkarthik Kalyanasundaram of Cupertino CA (US)
- Guru Prakash Arumugam of Sunnyvale CA (US)
- Srinivas Vasudevan of Mountain View CA (US)
- G10L15/22
- G06F3/16
- G10L15/16
- G10L15/18
- G10L15/30