#aight im here
1 messages · Page 1 of 1 (latest)
provided with ha on a laptop that runs ha os
ok
still whackin at this thing,
is there a way i can help
actually
I just had to re-read an article...
"I managed to make it work by changing the file name so that it matches the dataset name in the json file, e.g. this is my (Italian) giulia.onnx.json:"
change the name of the files in /share/piper to jarvis.onnx and jarvis.onnx.json
restart piper and it'll work
there are some minor changes that I would make, though instead of just changing the file names
(it'll work, but if you have the same 'voice' in different pitches, it will get muddy)
it still doesnt work for me, did u make a piper folder urself ? and can u see the model in the piper config ?
I used filezilla to create /share/piper
piper configuration, no. however, in the voiceassistant configuration it is an option
basically, here's my step-by-step ....
when i check it there it stil show medium but the names are changed to jarvis
arent*
create folder /share/piper on HAOS
downloaded onnx and onnx.json files, renamed to "en_GB-jarvis-high.onnx, en_GB-jarvis-high.onnx.json"
edit en_GB-jarvis-high.onnx.json, changed "dataset": "jarvis" to "dataset": "en_GB-jarvis-high"
uploaded to /share/piper on HA Host
HA Settings -> Add-ons -> Piper
Stopped service, started service
HA Settings -> Voice Assistants -> Your VA
Text-to-speech language -> en_GB
voice -> jarvis (high)
[update]
👌
i did everything u said and i get this
en_GB-jarvis-medium (medium)
my files are named
"en_GB-jarvis-high.onnx, en_GB-jarvis-high.onnx.json"
and the name of the dataset has been changed to
en_GB-jarvis-high
Where are you seeing medium?
bcs i have the medium and the high i must have used the medium on that one but doesnt matter bcs both are changed like instructed
(Like, which window)
in assist pipeline
Ahh. Ok so are they both in the list now?
yes they where before also but didnt play
it needs dubble .onnx ?