TTS Web UI
Model + speed dropdowns. Generates WAV for preview + download.
Single-file app.py
Text
Tip: leave punctuation in; it helps prosody. App adds tiny pauses after sentences automatically.
Model
Estonian VITS (Common Voice)
English VITS (LJSpeech)
Speed
Very Slow (1.45)
Slow (1.35)
Calm (1.30)
Slightly Slow (1.20)
Default-ish (1.10)
Generate
Reset
What “Speed” does:
it changes
length_scale
. Higher = slower.
Note:
if you want even calmer, try 1.35–1.45.
Output
Nothing generated yet. Hit
Generate
.
Status:
waiting
This app keeps only the last generated WAVs in
/tmp/tts_webui
.
Ops notes
Run behind your reverse proxy if exposing publicly.
Consider Fail2ban on the proxy if you publish it.
Models are loaded on demand and cached in memory.
Output dir:
/tmp/tts_webui