Festival Speech Synthesis and Codec2

Here'a a bash script that takes as its first argument a string of quoted text. It then passes this through the festival text-to-sound processor and plays two versions of the result. The first version is unprocessed; the second version has been encoded with codec2 and then decoded.

The results are generally impressive. But I hope this might be some help in improving the codec. For instance, the roughening of sibilants like 'sss' in English (which becomes something like a 'sh' under codec2) is easily reproduced with festival.

To use this script, change the first line to point to the directory in which your c2enc file resides, save the script as, say, 'tts-codec2.sh' and then issue the command 'chmod a+x tts-codec2.sh' so that the script is executable. Then run it on the command line with 'tts-codec2.sh "Salmon steaks grill beautifully."' You'll hear the non-codec2 version, then the codec2 version.

CODEC2_HOME=/home/brucerob/Codec2/codec2/src
TFILEA="/tmp/$(basename $0).$$a.tmp"
TFILERAW="/tmp/$(basename $0).$$.raw"
echo $1 | text2wave | sox - -r 8000 -t raw -s -2 $TFILERAW 
play -r 8000 -t raw -s -2 $TFILERAW
$CODEC2_HOME/c2enc $TFILERAW $TFILEA
$CODEC2_HOME/c2dec $TFILEA $TFILERAW
du -sh $TFILEA
play -r 8000 -t raw -s -2 $TFILERAW

Comments

Post a Comment

Popular posts from this blog

Building and Operating a Transmitter from the Roaring 20s

Image
Radio was booming in the 1920s: transmitters had started using the pure tone of 'continuous wave' signals, replacing signals sent with sparks, and increasingly selective and sensitive receiver circuits offered greater range seemingly by the month. Amateurs, not yet able to tame their simple transmitters with crystal-controlled oscillators, constructed them on simple wooden boards with heavy wiring and large inductor coils in order to keep heating from causing their signals to drift. These homebrew transmitters were ongoing experiments in electronics and engineering. The position of a coil might increase output power but cause the signal to sound like a buzzsaw. If you could find a higher voltage power supply, that too would raise its power, but the increased currents in the rig would add more drift: if the other guy can't find your signal wandering up and down the band, what use is that extra 3 watts? The 1929 QSO Party Every year the Antique Wireless Association holds a tw
Read more

Defining a 49:1 EFHW Transformer's Losses

Image
 End-fed halfwave antennas are so convenient, especially in rapidly deployed setups, and they're easy to build . But how much energy is lost in the 49:1 transformer that makes it possible? Because I operate mainly QRP, and I'm using a rather large core for 5w, I'm not concerned with losses that appear once the transformer's core reaches the Curie temperature. So I can use a nanoVNA, performing a S21 logmag trace on two cores back-to-back, to see how much power my transformer is eating up on the way to the antenna. I used the Fair-Rite 2643625002 core, inspired by excellent results MM0OPX reports in a generously shared spreadsheet of results. Could I get around 0.5 dB loss from 80m all the way to 10m if I used that core in 21:3 turns ratio?  My experience with more donut-shaped ones suggested that such broad-banded performance was a tough order: a 14:2 turns ratio works well on one end of the HF band; 21:3 works well on the other. In the end, Colin's accompanying vi
Read more