Re: DTMF v4

On 2012-12-14 23:51, Roman Shpount wrote:
> Tones of at least 40 ms in duration with 40 ms pauses should be 
> recognized. Part of the reasoning is that loosing a single audio 
> packet (typically 20 ms) would still allow DTMF tones and gaps between 
> them to be correctly recognized.

Good point.
I saw a recommendation for 65 tone / 65 silent gap for production, and 
40/ 40 for detection in some private blog. That can be just because of 
the packet loss reasoning.

Gunnar

Received on Friday, 14 December 2012 23:34:42 UTC