Feature request: checking watermark of a stream in real time

[Irina R]

Hi,

I wonder if it would be possible to write progressively precise watermark information on stdout as the stream is coming in. This would make for a nicer demo, where the audience doesn't have to sit through the whole 5 minutes of recording and could see some results earlier. Just a nice-to-have.

[Stefan Westerfeld]

Hi!

Am 20.04.22 um 17:35 schrieb Irina R:

First of all, the watermark is organized in data blocks (of about 50
seconds), where each data block contains all the 128 data bits, but not
in linear order. Which means that for a possible streaming decoder, you
will not see the bits coming in one by one, but approximately every 50
seconds, after some buffering, a whole 128 bit message would be decoded.

It is not feasible/practical to simply make everything we currently do
in 'audiowmark get' streamable, some things are probably better left as
they are, but also rely on having the whole file available.

So we'd have to add a streaming decoder as a extra new feature like
'audiowmark get-stream', which is a bit of work. It has not very high
priority for me to add it, if there is no other use case than "nicer
demos". For this case, you could get something similar by simply by
writing a script which

repeats (forever):
- record 50 seconds as foo.wav
- run audiowmark get foo.wav

Cu... Stefan
--
Stefan Westerfeld, http://space.twc.de/~stefan