System and methods provide acoustic echo monitoring and cancellation for real time media processing in an internet protocol (IP) media server in an IP network. An echo monitor is configured to selectively compare audio streams into and out of the IP media server through a selected port. The comparison determines an occurrence of an echo. An echo canceller in communication with the echo monitor is configured to respond to the determination by the echo monitor so as to remove the echo from at least one of the audio streams. A talk burst detector may be used to detect speech in at least one of the audio streams through the selected port. The echo monitor selectively compares the audio streams in response to a signal from the talk burst detector that indicates detection of speech.