摘要 |
A determination to enable echo cancellation is made in advance of a real time communication session. A model sample audio file (214) stored in memory on the computer system (200) is played through a loudspeaker (204) connected with the computer system (200). The sound generated by the loudspeaker (204) is captured by a microphone (202) connected with the computer system (200) to create a captured audio signal. The captured audio signal is correlated with the sample audio signal to determine the presence of any echo effect in the captured audio signal (414). Characteristics of any echo in the captured audio signal, for example, frequency, delay, and gain, are parameterized (418) and stored as initial values (420) for use by the echo cancellation algorithm. The stored parameters are used by the echo cancellation algorithm at the initiation of a communication session (604) until the algorithm receives enough real time information to make any necessary ongoing adjustments (608). |