Fact · Voice and AI
1.6seconds of speech per second
Home Assistant says that on a Raspberry Pi, using medium quality Piper models, the text-to-speech engine can generate 1.6 seconds of voice in one second.
Checked against its source on October 17, 2026
In the source's own words
“On a Raspberry Pi, using medium quality models, it can generate 1.6s of voice in a second.”
Installing a local Assist pipeline, Home Assistant project, Home Assistant, 2026, Text-to-speech engine.
Open the original- What it measures
- This is text-to-speech speed, the time to speak a reply, and it is a different measure from the time to understand a spoken command.
- What it does not show
- A project-reported figure for medium quality models on a Raspberry Pi. Other models and hardware will differ.