Echo detection

An echo detection system comprising:

• data processing hardware;
• and
• memory hardware in communication with the data processing hardware, the memory hardware storing instructions that when executed on the data processing hardware cause the data processing hardware to perform operations comprising:
• receiving a frequency representation of a playout audio signal corresponding to a microphone audio signal;
• determining features based on the frequency representation of the playout audio signal, each feature corresponding to a pair of frequencies of the frequency representation of the playout audio signal and a period of time between the pair of frequencies, each corresponding pair of frequencies comprising a respective first frequency and a respective second frequency, the respective first frequency different than the respective second frequency;
• determining that a match occurs between a first feature based on a frequency representation of the microphone audio signal and a second feature based on the frequency representation of the playout audio signal;
• and
• determining that an echo within the microphone audio signal exists when a delay value between the first feature and the second feature satisfies an echo value threshold that is predictive of a respective echo being human perceptible.