I get the error: Failed to perform the action tts.cloud say. Please call connect first.
same happens with tts.home_assistant_cloud
I consulted an AI about this. It says:
🧨 Why this suddenly started happening
This is the part that trips people up.
A recent HA update changed the behavior of Assist-based media players:
• They no longer auto-connect for TTS
• They only connect during an active Assist session
• TTS to these entities is no longer supported unless the device explicitly supports standalone playback
This is why it “used to work” and now doesn’t.
You didn’t break anything — the underlying behavior changed.
Is that true or not and if it is, is there a fix for it?
(AI says not to use it… That’s not a fix! I use my voice preview extensively for TTS and I would very much like to keep doing it…)
As Far as I know nothing has changed with TTS to media players.
So if you want help you need to provide more info. How did you configure your VPE, what hardware and software you are using for TTS and anything else that may be relevant.
FYI, I did my due diligence first including: researching the web, the community, Github, asking AI, rebooting, reloading, all that before I even thought about posting something here… the only thing I did find is the above AI diagnostic I got from Chat GPT…
Everything was working fine before. I didn’t change anything recently save for updating HASS core and integrations to their latest iterations and then voice preview suddenly stopped doing any and all TTS from my automations with the above “please call connect first” error. It does that whether I use home assistant cloud or Google TTS btw.
I dunno: maybe I’m the only one with this problem? if that’s the case, I’ll open an issue…
As for additional info, I’m using Home Assistant Green with a Nabu Casa account. the TTS hardware is the VPE (don’t understand what other hardware you might refer to?). My HASS Core and integrations are almost all up to date.
No, I think I am having the exact same issue. After updating Core, ESPHome, and i think the voice preview speakers also got an update, I stopped getting TTS from them. They wake up correctly, STT correctly, and get a response from HA+TTS, but they never play it. Assistant works perfectly on all other devices, including other ESP32 devices like the s3 box and atom echo. I’m running HA in a PVE VM and have tried a bunch of TTS options as well, like Piper, Speakers, HA Cloud, etc. Next up would be to roll one of the voice preview boxes back a version to see if that helps, i guess, will look into that when i can.
Home Assistant Core 2025.12.x forces TTS audio files to be generated in FLAC format, completely ignoring the preferred_format settings. The Voice Preview Edition speakers cannot successfully receive FLAC files through the ffmpeg_proxy service, resulting in silent playback timeouts.
I tried downgrading but ended up with db issues, so back to the new version for now and living with the silent speakers.
