Speaker recognition based on an inside microphone of a headphone
AssigneeApple Inc.
InventorSorin V. Dusan
Life of this patent
Estimating the term…
Abstract
A speaker recognition algorithm is trained (one or more of its models are tuned) with samples of a microphone signal produced by an inside microphone of a headphone, while the headphone is worn by a speaker. The trained speaker recognition algorithm then tests other samples of the inside microphone signal and produces multiple speaker identification scores for its given models, or a single speaker verification likelihood score for a single given model. Other embodiments are also described and claimed.
Claims31
A method for speaker recognition used with a headphone, the method comprising:
- obtaining a first microphone signal, produced by an inside microphone, and a second microphone signal, produced by an outside microphone, wherein the first microphone signal and the second microphone signal contain speech of a user, and the inside microphone and the outside microphone are integrated within the headphone;
- identifying or verifying the user based on the speech of the user using the first microphone signal, and not the second microphone signal, as input to a speaker recognition algorithm;
- and
- recognizing the speech of the user using the second microphone signal as input to a speech recognition algorithm.
A headphone comprising:
- a headphone housing having integrated therein an inside microphone, and one or more outside microphones;
- and
- a processor and memory having stored therein instructions which when executed by the processor, performs the following
- obtaining a first microphone signal, produced by the inside microphone, and a second microphone signal, produced by one of the one or more outside microphones, the first microphone signal and the second microphone signal containing speech of a user,
- identifying or verifying the user based on the speech of the user using only the first microphone signal as input to a speaker recognition algorithm, and
- recognizing the speech of the user using the second microphone signal as input to a speech recognition algorithm.
A smartphone or tablet computer having one or more processors configured to perform the following:
- obtaining a first microphone signal, produced by an inside microphone of a headphone, and a second microphone signal, produced by one of one or more outside microphones of the headphone, the first microphone signal and the second microphone signal containing speech of a user;
- identifying or verifying the user based on the speech of the user using only the first microphone signal as input to a speaker recognition algorithm;
- and
- recognizing the speech of the user using the second microphone signal as input to a speech recognition algorithm.
Dependencies are inferred from the claim text.