khimaros

i've written up a short tutorial here: https://khimaros.com/articles/docker-compose-podman/

note that this will not work well if rainloop is behind a reverse proxy due to https://github.com/RainLoop/rainloop-webmail/issues/1849

this workaround should also work for the reverse proxy case mentioned above: https://github.com/RainLoop/rainloop-webmail/issues/1849#issuecomment-742718083

I would be quite happy with an "import wav, convert to FLAC, write metadata" approach. Would this be a lower friction route than full support for non-metadata storing formats?

@sampsyo -- if someone were interested in working on this functionality, do you have design guidance or pointers for where to start looking?

here is an example of [streaming transcription from microphone in python](https://github.com/mozilla/DeepSpeech-examples/blob/r0.9/mic_vad_streaming/) with deepspeech

a few resources since i was just researching this: - https://github.com/castorini/howl - https://github.com/seasalt-ai/snowboy - https://github.com/Picovoice/porcupine you mention porcupine is not open source, do you mean the models?

if you are considering vosk, i'd recommend taking a look at [mozilla deepspeech](https://github.com/mozilla/DeepSpeech) which i found to be both easier to configure and considerably more accurate. they also have examples...

@Sispheor -- that's right, including vosk. i'm not sure if either is a good fit for a snowboy replacement, but if you're going to go that direction i think deepspeech...

snips looks promising but appears to have been acquired by Sonos. i couldn't find the platform source https://github.com/snipsco?q=platform&type=&language=&sort= and their usage docs have been removed. the closest available is https://github.com/snipsco/snips-record-personal-hotword