Feedback elimination method and apparatus

A method for detecting a singing frequency of audio in a room represented by an audio signal in a signal processing system, the method comprising:

• 1) maintaining a hit set of data points and a change set of data points for frequency domain energy bins in each time frame signal of the audio signal, wherein each data point has an energy strength and a frequency;
• 2) updating the hit set and the change set with frequency domain energy bins for each time frame signal of the audio signal;
• 3) repeating step 2) for a first number (It) of times;
• 4) determining a singing frequency of the audio signal which is a hit frequency with a maximum energy strength if the maximum energy strength exceeds a hit threshold and the strength of the change data point with the hit frequency exceeds a change threshold;
• 5) repeating steps 2-4 if a singing frequency of the audio signal is not determined;
• and
• 6) providing a singing frequency signal to allow suppression of the singing frequency.

A microphone module comprising:

• a microphone element for receiving audio in a room and providing an audio signal;
• and
• a processor coupled to the microphone element;
• wherein the processor is operable to perform a method of:
• 1) maintaining a hit set of data points and a change set of data points for frequency domain energy bins in each time frame signal of the audio signal, wherein each data point has an energy strength and a frequency;
• 2) updating the hit set and the change set with frequency domain energy bins for each time frame signal of the audio signal;
• 3) repeating step 2) for first number (It) of times;
• 4) determining a singing frequency is a hit frequency with a maximum energy strength if the maximum energy strength exceeds a hit threshold and the strength of the change data point with the hit frequency exceeds a change threshold;
• 5) repeating steps 2-4 if a singing frequency is not determined;
• and
• 6)providing a singing frequency signal to allow suppression of the singing frequency.

An audio system comprising:

• a microphone element for receiving audio in a room and providing an audio signal;
• a loudspeaker for providing audio to a room;
• an amplifier coupled to the microphone element and the loudspeaker;
• and
• a processor coupled to the microphone element and the amplifier;
• wherein the processor is operable to perform the method of:
• 1) maintaining a hit set of data points and a change set of data points for the frequency domain energy bins in each time frame signal of the audio signal, wherein each data point has an energy strength and a frequency;
• 2) updating the hit set and the change set with frequency domain energy bins for each time frame signal of the audio signal;
• 3) repeating step 2) for a first number (It) of times;
• 4) determining a singing frequency of the audio signal which is a hit frequency with a maximum energy strength if the maximum energy strength exceeds a hit threshold and the strength of the change data point with the hit frequency exceeds a change threshold;
• 5) repeating steps 2-4 if a singing frequency is not determined;
• and
• 6) providing a singing frequency signal to the amplifier to allow suppression of the singing frequency.