ComfyUI Node
Antalia-2 Mini Speak (Turkish TTS)
Speaks Turkish text with Antalia-2 Mini's one synthetic male voice. Raw text is fine: numbers, dates, times, amounts, initialisms, web addresses and English or brand words are normalised by the package. The same text, seed and settings give the same audio.
Antalia-2 Mini Speak (Turkish TTS)
- model
- audio
- spoken_text
- report
◄textMerhaba, size nasıl yardımcı olabilirim?►
◄seed0►
◄speed0.95►
◄steps8►
◄cfg2.0►
◄sample_rate48000►
◄eqtrue►
Categoryaudio/antalia
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| model | ANTALIA_MODEL | — | |
| text | STRING | Merhaba, size nasıl yardımcı olabilirim? | — |
| seed | INT | 00–2147483647 | — |
| speed | FLOAT | 0.950.5–2 | speaking rate; 0.95 is the release default (slightly slower than the voice's own pace) |
| steps | INT | 81–32 | flow-matching steps; 8 is the release default, 4 is faster and measurably less accurate |
| cfg | FLOAT | 2.01–5 | classifier-free guidance; 1 turns it off |
| sample_rate | COMBO | 48000 | 48000 is native; 24000 / 16000 / 8000 (telephone) are resampled by the package |
| eq | BOOLEAN | true | the release's output EQ and de-esser; off gives the raw vocoder output |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| audio | AUDIO | — |
| spoken_text | STRING | — |
| report | STRING | — |