I flashed a echo dot gen2 with echomuse and I’m now looking for a local pipeline that runs completely on a cpu.
2 years ago I started using vosk but stopped when my speaker died.
Is vosk still the way to go when pre-defining sentences? I’m a bit put off by the repo only having had one commit in the last 2 years (and that was to add a model).
rhasspy-speech got deprecated too. Which makes me wonder whether I’m looking for the wrong things?
Do I have to accept that I absolutely need a GPU now?
Any help is appreciated! Thanks 