A voice processing apparatus including a memory, and a processor coupled to the memory and the processor configured to acquire a first input signal containing a first voice, and a second input signal containing a second voice, obtain a first signal intensity of the first input signal, and a second signal intensity of the second input signal, specify a correlation coefficient between a time sequence of the first signal intensity and a time sequence of the second signal intensity, determine whether the first voice and the second voice are in the conversation state or not based on the specified correlation coefficient, and output information indicating an association between the first voice and the second voice when it is determined that the first voice and the second voice are in the conversation state.