Hey Patrick, Many thanks for your tests, I warmly appreciate.
Indeed I updated to 3.4 and it kind of worked again, except for two things. A few months ago I managed to configure my system so that PulseAudio runs only once and all the speech-dispatcher servers connect to the same instance of PulseAudio. That was convenient because even the instance of BRLTTY that renders the text consoles was able to talk, so I was in a state where I had speech everywhere and found that very nice, it made me faster and less tired than reading in braille only as I did for years. This is now broken. Also, since my system stopped working I thought that I had nothing to lose in migrating from PulseAudio to pipewire, which I did. So now speech-dispatcher is using pipewire to output sound and I am under the impression that the quality of the voice has decreased compared to before. am not sure whether this is due to switching from PulseAudio to pipewire or whether it can come from the upgrade of the Vocalizer voice but I assume that the upgrade is only on the packaging and does not change the binaries of the TTS themselves... Apologies for the long story, Seb.
