dwks

Try vad_padding_end_ms in kaldi_module_loader_plus.py. I myself have set it to 200ms up from 100 to avoid this problem.

Yes, this works. Please consider adding this to the README, it's pretty confusing to start using kaldi-active-grammar on Linux currently. P.S. `pip install dragonfly2[kaldi]` pulls in `kaldi-active-grammar` automatically.

If you don't want to mention dragonfly on the kaldi-active-grammar page, could at least mention it here: https://github.com/daanzu/kaldi-grammar-simple

For dragonfly, I think it would be something like this for "blah" or "dictate blah": ``` from dragonfly import * class DictationRule(CompoundRule): spec = ('[dictate] []') extras = [Dictation(name='dictation')] exported...

Thanks to the rowhammer attack, it looks like Linux now requires root privileges to access /proc/pid/pagemap (even if you ran the process yourself): > To make life harder for attackers...

@Zibri, can you confirm exactly which architecture you are using? arm32 vs arm64. Thanks.

Thank you for such a detailed message! Indeed, the [COUGH] etc tokens should be ignored by the silvius parser otherwise you get tons of non-recognitions. You need this change https://github.com/dwks/silvius/commit/67bb67ad6f5e4b40ca48d564dc48b7a0d6d48d89...

Sorry for the delay. I'm not familiar with that microphone, it looks decent. The speech model is probably having trouble with your voice, it may be your accent, or other...

This is a tricky problem, to programmatically generate the doc strings. I like your solution, though eventually it would be best to not instantiate the grammar twice of course. Before...

P.S. Some command words, like "comma" or "ctrl", should be allowed in an english context. For example: "sentence hello world bang control sierra". At least, this is how I envisioned...