#Google Cloud TTS voice error

15 messages · Page 1 of 1 (latest)

modern marsh
#

I had this action working for a while, but at some point I can't be sure of, it started throwing some errors.

action: tts.speak
target:
  entity_id: tts.google_cloud
data:
  cache: true
  media_player_entity_id: media_player.google_mini_speaker,media_player.google_nest_speaker
  message: >-
    Hoy toca pollo.
  options:
    voice: es-ES-Chirp-HD-F

Error message attached as it's quite lengthy, because it lists all possible voices. Funny thing is that the one chosen in the action is in that list. So I don't get why the error.

Any idea?

📎 message.txt
drifting sluice
#

Try

'es-ES-Chirp-HD-F'

#

Copy it right from that error message in case of some weirdness i gont or something.

modern marsh
#

It happens with any voice as well. I've tried copying some of the others in the list

drifting sluice
#

That's from my blueprint for doing tts. It's a bit dated, but noone has told me it's broken.

modern marsh
# drifting sluice You are picking voice, you might want to try some of the other options, maybe it...

thanks, just tested it, and it's using the legacy version of google cloud, which I cannot use since I removed the yaml long ago. https://www.home-assistant.io/integrations/google_cloud/#action-say-legacy

So I get

Logger: homeassistant.components.script.tts_say_message_2024_06_08
Source: helpers/script.py:2039
integration: Scripts (documentation, issues)
First occurred: 10:26:27 (4 occurrences)
Last logged: 10:26:49

TTS_say_Message-2024-06-08: Main TTS Choose Statement: choice 4: Error executing script. Service not found for call_service at pos 1: Action tts.google_cloud_say not found
TTS_say_Message-2024-06-08: Error executing script. Service not found for choose at pos 2: Action tts.google_cloud_say not found
#

Found the issue

#

it wasn't enough with setting the voice at the action level

#

I was even removing all options in the action and still getting the rror

#

which made me think maybe the error was triggering from the default settings set at the integration level

#

indeed

#

I had language and gender, but for some reason it forces you to fill a voice as well, otherwise it will trigger the error

#

quite weird, docs could be updated I think