Nvidia patents car voice assistant that checks manuals before answering
Nvidia describes a system that listens to questions about a vehicle and finds the most relevant parts of the owner's manual or past answers. It combines this information into a prompt for a language model to generate a specific response.
In the filing’s own words
In various examples, systems and methods that use dialogue systems associated with various machine systems and applications are described. For instance, the systems and methods may receive text data representing speech, such as a question associated with a vehicle or other machine type. The systems and methods then use a retrieval system(s) to retrieve a question/answer pair(s) associated with the text data and/or contextual information associated with the text data. In some examples, the contextual information is associated with a knowledge base associated with or corresponding to the vehicle. The systems and methods then generate a prompt using the text data, the question/answer pair(s), and/or the contextual information. Additionally, the systems and methods determine, using a language model(s) and based at least on the prompt, an output associated with the text data.
What exactly they claim (claim 1)
1. A method comprising: receiving, using one or more microphones of a machine, audio data representative of a question associated with the machine; generating, based at least on the audio data, a first embedding representative of the question associated with the machine; determining, based at least on the first embedding and one or more second embeddings associated with one or more question-and-answer pairs, at least one or more scores indicating one or more similarities of the one or more question-and-answer pairs to the question; determining, based at least on the first embedding and one or more third embeddings associated with a manual corresponding to the machine, contextual information that includes one or more portions of the manual that are relevant to the question; generating prompt data for one or more language models that is representative of at least first text associated with the question, second text associated with the one or more question-and-answer pairs that are arranged based at least on the one or more scores, and third text associated with the contextual information; applying the prompt data to the one or more language models; generating, based at least on the one or more language models processing the prompt data, output data corresponding to a response associated with the question; and outputting, using one or more components of the machine, content associated with the response.
Facts
- Company
- Nvidia (NVIDIA Corporation)
- Official title
- Dialogue systems using knowledge bases and language models for automotive systems and applications
- Patent number
- US12801859B2
- Application number
- 17947491
- Filed
- Sep 19, 2022
- Granted
- Nov 17, 2026
- Inventors
- 7
- Claims
- 22
Classified under
G06F 40/35 Physics & computingMore from Nvidia in Cars & transport
- Nvidia patents robotaxi system with AI dispatcher and virtual rails Sep 15, 2026
- Nvidia patents car drowsiness check that uses your body clock Oct 13, 2026
- Nvidia patents 3D camera system to measure passengers in vehicles Oct 13, 2026
- Nvidia patents cruise control that predicts driver's next move Oct 20, 2026
- Nvidia patents system to rate how much a car's sensors can trust their own data Nov 3, 2026
Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.
