I have voice data at 128 kbps coming from three different sources. I'm looking for mixing all three voices together and sending it back to speakers. What methodologies are there for mixing digital audio bit streams and what will be the mixed voice data rate? will it be 128*3 or just 128 only?
When you add multiple audio sources, you must first make sure that they use the same sampling rate, and then same numerical format, so that addition can be realized.
So, if they are not at the same sampling rate already, then you have to resample them to the same rate. And also make sure (by type conversions) that all of them to use the same numerical data format (same type and precision, etc.)
When you say 128 kbps alone; it looks like either an compressed data stream such as mp3, or PCM raw data at 16 kHz, 8-bits per sample. Assuming the latter is the case, then the resultant stream will again be an 16 kHz, 8-bits per sample PCM data stream, having a bit rate of the same 128 kbps.