I am using an automation to get data and process back for agent response to then feed to tts.piper and play it on an Apple HomePod Mini. This worked until recently and I am running the latest everything but here is what I am getting in the errors:
Executed: April 28, 2026 at 6:30:04 AM
Error: ('failed to init decoder', -1)
Result:
params:
domain: tts
service: speak
service_data:
cache: false
media_player_entity_id: media_player.great_room_2
message: >-
Today is Tuesday April 28, 2026. The weather is rainy across the area.
Today's high temperature will be 76 degrees Fahrenheit and tonight's low
temperature will be 63 degrees Fahrenheit. There is a chance of rain with
a probability of 48 percent.
options:
voice: en_GB-alan-medium
entity_id:
- tts.piper
target:
entity_id:
- tts.piper
running_script: false
Here is how I am calling this action:
action: tts.speak
metadata: {}
data:
cache: false
media_player_entity_id: media_player.great_room_2
options:
voice: en_GB-alan-medium
message: "{{ agent.response.speech.plain.speech }}"
target:
entity_id: tts.piper
Trying to figure out what the issue is and appreciate any insight. Thanks!