P
Peadar O'Colmain
Guest
Hi all, Peadar in Ireland again.
I know I thought that I had cracked this issue but although I can get the Acoustic Echo Cancellation Routine to do its job and send the sound to another computer and play it fine, I am still stuck trying to compress and decompress that same voice string with ffmpeg/H264.
I can compress my voice from the microphone and send it over through a holepunched socket and then decompress and play it. That works really well. When I try to do exactly the same thing with the output from the AecSDkDemo the voice is just a bubbly sound at the receiving end. I am using exactly the same code in both programs to compress, send, receive, decompress and play. I have examined the string values in both cases, those from the microphone and from the AecSDKDemo. They are both composed of bytes between -127 and 128. Neither ever goes outside that range, yet the voice from the mike will go over fine and the voice from the AecSDKDemo is still dreadful. I have made sure they both have plenty of extra memory. Any help will be greatly appreciated and and solution will certainly be made public here. Thank you all in advance. Peadar
Continue reading...