More emphasis on digital voice quality

[quad_track]

Experimenting has always been a mainstay of amateur radio. When talking about digital radio, I believe there are two main directions to push into: go into low SNR, weak signal teritorry and FreeDV already does that very well with it's OFDM, LDPC encoded modes which compete against SSB; the other direction is to try to compete with narrow-FM on quality and SNR levels. For the experimenter, using components that allow the maximum flexibilty and let us tweak the inner working is a must. That usually means using only free and open source software. Advances in digital voice codecs are happening every day so this allows us to push more in one direction or another.

Here is my take on this topic:

So start with a high bitrate (10 kbit/s) codec, fit into 12 kHz channel, and then begin to lower the bitrates in pusuit of lower SNR but preserve 90% of the quality. Interested parties, keep an eye on advances with SILK on the Xiph.org mailing lists.

 

[FeedForward]

I can't say what your point is, especially after listening to the example. Are you asking experimenters to come up with better digital voice quality? If so, relax. We already have the ability to transmit and recover a signal with any arbitrary degree of accuracy. Whew! Hams can sound like they are reading the news on NPR, over even better! Certain sectors of radio experimentation, such as amateur, tend to focus on intelligibility rather than on distortionless fidelity. In fact it can easily be shown that intelligibility and fidelity do not always go hand in hand...Then our discussion, not limited to amateur radio, must include the available bandwidth to achieve a given level of accuracy. On this planet, physics has to follow that rule. Nothing against experimentation here. New Codecs, new error correction schemes, better methods for storing information are all great. What I don't understand is the niche group of amateurs who think that developing traditional radio communication skills is not a valid pursuit. Listening to a signal through noise, distortion, frequency instability - the whole lot - is somehow no longer necessary or fun.