Well i tried the new version from tschmidty69, but it doesn’t work. It recognises the hot word but then i get:
INFO:snips_hotword_hermes : Hotword detected: site_id:default model:hey_snips
[12:13:54] [Hotword] detected on site default, for model hey_snips
INFO:snips_dialogue_lib::coordinator::coordinator: Site [default]: reset site -> turn off ASR and turn off Hotword
INFO:snips_dialogue_lib::coordinator::coordinator: Session [ee7e9393-f625-4985-bf6b-b9be873e61d5]: The session has been created because the site "default" is free
[12:13:54] [Asr] was asked to stop listening on site default
[12:13:54] [Hotword] was asked to toggle itself 'off' on site default
[12:13:54] [Dialogue] session with id 'ee7e9393-f625-4985-bf6b-b9be873e61d5' was started on site default
[12:13:54] [AudioServer] was asked to play a wav of 41.1 kB with id '51855bb3-48a5-46b7-953d-c8031558715b' on site default
INFO:snips_audio_portaudio : Playing "51855bb3-48a5-46b7-953d-c8031558715b" using output "default", wav spec: WavSpec { channels: 2, sample_rate: 22050, bits_per_sample: 16, sample_format: Int }
INFO:snips_hotword_lib::features : Audio thread for site default idle.
INFO:snips_audio_portaudio : Playing of "51855bb3-48a5-46b7-953d-c8031558715b" finished
[12:13:54] [AudioServer] finished playing wav with id '51855bb3-48a5-46b7-953d-c8031558715b'
WARN:snips_tts_hermes: Did received a "play finished" message from audio server with an unknown id '51855bb3-48a5-46b7-953d-c8031558715b'. It could be normal if you have more than one audio server.
[12:13:54] [Asr] was asked to listen on site default
INFO:snips_asr_hermes : Listening
INFO:snips_asr_lib::asr : T0 entered AsrRunner::run
INFO:snips_asr_lib::asr : T0+0.000 capture started
1544267634: Socket error on client snips-asr|137-local-snips-lb-1, disconnecting.
1544267634: Socket error on client snips-asr|137-local-snips-lb-3, disconnecting.
2018-12-08 12:13:54,858 INFO exited: snips-asr (terminated by SIGILL (core dumped); not expected)
2018-12-08 12:13:55,860 INFO spawned: 'snips-asr' with pid 147
1544267635: New connection from 127.0.0.1 on port 1883.
1544267635: New client connected from 127.0.0.1 as snips-asr|147-local-snips-lb-1 (c0, k10).
1544267635: New connection from 127.0.0.1 on port 1883.
1544267635: New client connected from 127.0.0.1 as snips-asr|147-local-snips-lb-3 (c0, k10).
Is it me or does this not work?
I have the latest version of the add on and downloaded a new assistant from the console
ERROR:snips_nlu : Unable to read file ‘/usr/share/snips/assistant/nlu_engine/nlu_engine.json’
→ caused by: Expected model version 0.16.0 but found 0.17.
Hi TinyDoT! I’m trying to get snips setup. I have the snips mqtt broker on my Pi and the mqtt broker in hassio bridged and connected to each other. When I tell snips to turn the lamp on, it recognizes my request and the correct intent. BTW, my assistant has the default HomeAssistant app from the snips store. Here is what I see using “sam watch” on the Pi: https://imgur.com/a/A4saaRo. So it seems hass is either not receiving the mqtt message or it’s getting it but not processing it properly. Can you provide some guidance? Thank you
EDIT: I have “snips:” added to my configuration.yaml but I keep getting this notification in the frontend: https://imgur.com/a/RLTs4Aw.
Running Hassio (docker install) on an Ubuntu server and Snips on a Pi.
EDIT: I pointed the Pi’s mqtt broker to hassio and I can see in the hassio logs that it connects successfully. However, I think maybe I need to configure the Pi’s snips.toml file so it’s looking at the correct broker. Can you confirm? If so, could you supply an example of your snips.toml?
From my understanding when you the HomeAssistant app from the Snips Store Snips and HomeAssistant need to be on the same Pi in order for it to work, your setup will not work with this app. To get your setup to work you need to use some of the other Apps and then create intents in the configuration.yaml under, intent_script:, something like this:
intent_script:
searchWeatherForecast:
speech:
type: plain
text: >
The temperature is currently,
{{ states('sensor.dark_sky_temperature') | round(0) }}
degrees outside, and {{ states('sensor.dark_sky_summary') }}.
The high for today is going to be,
{{ states('sensor.dark_sky_daytime_high_temperature_0') | round(0)}} degrees,
and, the low will be,
{{ states('sensor.dark_sky_overnight_low_temperature_0') | round(0) }} degrees.
I’m using part of the Weather App published from Snips for this. You have to uncheck “Only show apps with actions”. I hope this gets you started.
I’ve been using snips for a few days and the truth is that I think it’s a great component and although voice recognition (ASR) works very well for me, the feedback voice (TTS), in my case the voice in Spanish, is the part that doesn’t convince me.
Today I just realized that I can use google voices, but I do not mean the integration of snips with google (which is a paid service, by the way), I mean that when I define an attempt_script in HA, in the block action, I can use the service tts.google_say and send the answer to any of my multimedia devices.
For example:
intent_script:
DoorStatus:
action:
- service: tts.google_say
data_template:
entity_id: media_player.laptop
message: The door is {{ states('sensor.door') }}