Home Assistant voice users hear Markdown and model tags aloud
Home Assistant users sending LLM responses or alerts to text-to-speech can hear Markdown punctuation, such as asterisks, and sometimes model “thinking” tags spoken literally. Some speech engines normalize formatting, but users report that Piper and other setups do not; workarounds include changing prompts or patching Piper rather than adding a simple cleanup step before speech.
For home Assistant voice-assistant users. Mentioned from Feb 2025 to May 2025 on product forums.
5 different people described this problem in 3 separate discussions.
- Indie fit
- 3.0/10
- Pain
- 5.7/10
- Frequency
- 6.5/10
- Willingness to pay
- 0.0/10
- Momentum
- 5.0/10
- Who pays
- Consumers
- Competition
- Low
- Build difficulty
- Medium
What people said
Quoted word for word. Follow a link to read the whole discussion.
These characters are then said by the TTS (some tts engines do the normalisation themselves, but most can’t, Piper included). It would be a great idea to add a simple filter (as an option) to the assist pipeline (after the conversation module) that removes the formatting from the text
Got tired of waiting and added the changes to the piper myself Edit handler.py
Build brief
See what to build and who will buy it
- 1 product idea with the smallest useful version and pricing
- 3 places to find your first customers
- 4 more quotes from people who have this problem
- Current workarounds, existing solutions and risks