Back to results
US 2026/0301730 A1Published application
AI Voice
OwnerBao Traninventor
Life of this patent
Estimating the term…
Abstract
A computer-implemented method for an artificial intelligence voice agent system includes analyzing voice for intent recognition and sentiment analysis using a multimodal large language model (LLM); generating a response using one or more machine learning models; and converting generated response to synthetic speech through text-to-speech (TTS) engines while maintaining a sub-300 ms response time during an interaction.
Claims20
A computer-implemented method, comprising:
- analyzing voice for intent recognition and sentiment analysis using a multimodal large language model (LLM);
- generating a response using one or more machine learning models;
- converting generated response to synthetic speech through text-to-speech (TTS) engines while maintaining a sub-300 ms response time during an interaction.
Dependencies are inferred from the claim text.