Artificial intelligence device and operating method thereof
AssigneeLG Electronics Inc.
InventorHwansik YUN
Life of this patent
Estimating the term…
Abstract
An artificial intelligence (AI) device may acquire a probability that a received speech signal is classified as a noise signal, calculate a confidence level of a first model for determining to which phoneme the speech signal belongs, based on the speech signal, determine a weight of the first model based on the probability and the confidence level of the first model, and output a speech recognition result of the speech signal using the determined weight of the first model.
Claims20
An artificial intelligence (AI) device comprising:
- a memory;
- an output interface including a speaker or a display;
- a microphone configured to acquire a speech signal;
- and
- one or more processors configured to:
- determine a first probability that the acquired speech signal is classified as a noise signal,
- determine a confidence level of a first model based at least in part on the acquired speech signal, wherein the first model determines associations between corresponding phonemes to the acquired speech signal,
- determine a first weight of the first model based at least in part on the determined first probability and the determined confidence level of the first model, and
- output, via the output interface, a speech recognition result of the acquired speech signal using the determined first weight of the first model.
A method of operating an AI device, the method comprising:
- acquiring a speech signal;
- determining a first probability that the acquired speech signal is classified as a noise signal;
- determining a confidence level of a first model based at least in part on the acquired speech signal, wherein the first model determines associations between corresponding phonemes to the acquired speech signal;
- determining a first weight of the first model based at least in part on the determined first probability and the determined confidence level of the first model;
- and
- outputting a speech recognition result of the acquired speech signal using the determined first weight of the first model.
A non-transitory recording medium having recorded thereon a computer program for performing a method of operating an artificial intelligence (AI) device, the method comprising:
- acquiring speech signal;
- determining a first probability that the acquired speech signal is classified as a noise signal;
- determining a confidence level of a first model based at least in part on the acquired speech signal, wherein the first model determines associations between corresponding phonemes to the acquired speech signal;
- determining a first weight of the first model based at least in part on the determined first probability and the determined confidence level of the first model;
- and
- outputting a speech recognition result of the acquired speech signal using the determined first weight of the first model.
Dependencies are inferred from the claim text.