Hi All!
Today finally the M5Stack Atom Echo has arrived! 
So I installed it and it works even with Italian Language 
But it seems quite slow, and I don’t know if I can change something to get it better.
I have Home Assistant OS installed on Beelink MINI S12 Pro with Alder Lake-N N100, 16GB DDR4 +500GB M.2 PCIe 2280 NVMe. So I think it’s quite powerful!
I’m doing some tests and the logs on the whisper addon say:
INFO:faster_whisper:Processing audio with duration 00:02.560
INFO:wyoming_faster_whisper.handler: Accendi luce studio
Instead the Voice Assist debug has this for the wake word RAW which took 2,43 seconds:
entity_id: wake_word.openwakeword
metadata:
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
timeout: 5
wake_word_output:
wake_word_id: ok_nabu_v0.1
wake_word_phrase: ok nabu
timestamp: 1235
and this for the speech-to-text RAW which took 7,03 seconds:
metadata:
language: it
format: wav
codec: pcm
bit_rate: 16
sample_rate: 16000
channel: 1
stt_output:
text: " Accendi luce studio"
The Natural Language Processing took only 0,05 seconds:
conversation_id: null
device_id: 5051163320b288942a1c8a5c4096ebaf
intent_output:
response:
speech:
plain:
speech: Ho acceso
extra_data: null
card: {}
language: it
response_type: action_done
data:
targets: []
success:
- name: Studio
type: area
id: studio
- name: Studio
type: entity
id: light.studio
- name: Studio Strip
type: entity
id: light.studio_strip
failed: []
conversation_id: null
The settings for whisper is:
model: small-int8
language: it
beam size: 5
and this is all the whisper log:
s6-rc: info: service s6rc-oneshot-runner: starting
s6-rc: info: service s6rc-oneshot-runner successfully started
s6-rc: info: service fix-attrs: starting
s6-rc: info: service fix-attrs successfully started
s6-rc: info: service legacy-cont-init: starting
s6-rc: info: service legacy-cont-init successfully started
s6-rc: info: service whisper: starting
s6-rc: info: service whisper successfully started
s6-rc: info: service discovery: starting
[16:16:45] WARNING: Your CPU does not support Advanced Vector Extensions (AVX). Whisper will run slower than normal.
INFO:__main__:Ready
[16:16:48] INFO: Successfully send discovery information to Home Assistant.
s6-rc: info: service discovery successfully started
s6-rc: info: service legacy-services: starting
s6-rc: info: service legacy-services successfully started
INFO:faster_whisper:Processing audio with duration 00:02.560
INFO:wyoming_faster_whisper.handler: Accendi luce studio
Is there some wrong configuration? I think it takes a long time for the wake word and speech-to-text processing, I’m right?
Thanks in advance!