Skip to content
Nappai Documentation

Supertonic (local)

Supertonic (local) is one of the components that give a voice agent its voice: it turns the agent’s answers into audio for the caller to hear. It runs on the NappAI server: no credential, the text is not sent to any outside provider and there is no per-word cost. It has ten voices, five female and five male, that speak 31 languages. In Spanish it reads amounts, times, dates and phone numbers out in words. It is the recommended TTS to start with.

It plugs into the TTS connector of the Voice Agent node. It holds the voice settings the agent uses on every call.

When to use it

  • You are starting a voice agent and want it to speak without setting up any outside account.
  • You take calls in several languages and want the same voice in all of them.
  • You do not want the agent’s answers to leave the NappAI server.

Configuration

FieldWhat it does
VoiceThe voice: F1 to F5 are female and M1 to M5 male. Default: F2. All of them speak the caller’s language, which comes from the STT.
QualityHow much work the voice puts into each sentence, in steps. Fewer steps answer sooner and more steps sound cleaner; each step adds a little wait. Below 4, words can get slurred. Default: 5 steps. The 4 steps — Fast, 6 steps — Balanced and 8 steps — High quality options are reference points.
SpeedSpeaking pace. 1.0 is the voice’s natural pace; higher is faster and lower is slower. Accepts 0.5 to 2.0. Default: 0.95.

Output

The TTS output holds the voice settings. It connects to the TTS connector of the Voice Agent node, or to Fallback TTS when you turn on Fallback Engines and want it as a backup. The editor will not let you plug it into an STT connector by mistake.

Example

A hotel gets calls in Spanish, English and German and wants the same voice in all three languages.

  1. Add Supertonic (local) and connect it to the TTS connector of the Voice Agent.
  2. Pick a voice in Voice, for example F2.
  3. Use Parakeet (local, multilingual) as the STT with Detect Language on.
  4. Test the agent in the playground and listen to how it sounds.

Good practices

  • If the first sentence takes long to sound, choose 4 steps — Fast in Quality: it comes out sooner, at the cost of a little naturalness.
  • The differences between local and cloud voices are explained in Quality, noise and latency.