Hello Patrick, > David, Jean-Marc, would it be possible to backport the methods of the > 1200 bit/sec version to 2500 bit/sec and get even better quality on 2500 > bit/sec? I would say 2500 bit/sec is not yet toll quality, but maybe it > could be brought there.
I tried some clever tricks but stumbled across a simple one that works surprisingly well - I just changed the frame rate from 20 to 40ms. So at 1400 bit/s the amplitude information (encoded by LSPs) is updated every 40ms rather than every 20ms. I still update the pitch and voicing every 20ms at the lower rates. So it's not really a method we can use to increase the quality at say 2400 bit/s > <dream> > If some kind of hierarchical modulation could be used and the codec > supported both 1200 bit/sec and 2500 bit/sec, a communication system > could be designed, that would support good quality with a fallback when > channel properties get worse and data is not decodable. > </dream> Sure this is possible. But it would be nice to find some information we can add to make a dramatic increase in speech quality - as you point out the differences between 2400 and 1400/1200 are fairly small given we are halving the bit rate. At around 4000 bit/s we could start sending the phases of each harmonic, that would make a significant difference. Currently the phases are discarded, a we use a rule based approach to synthesise them at the decoder. This phase model is the cause of the "buzzy" artefact discussed a week ago by Ed. Cheers, David ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Freetel-codec2 mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freetel-codec2
