Skip to content
Nappai Documentation

Kokoro TTS

Kokoro TTS reads out the text you give it and returns a WAV audio file. It doesn’t need a credential or an account with an external service.

Use it when you want to turn a message, a summary or an agent’s answer into audio inside a flow.

When to use it

  • Send your customers a voice note confirming their appointment.
  • Turn the daily sales summary into audio to listen to on your way to the office.
  • Produce voice-overs for announcements or welcome messages.
  • Let an agent answer with audio, using the component as a tool.

Configuration

FieldWhat it does
Text to SpeechThe text to read out. You can type it or connect it from another component.
VoiceThe voice. Default: ef_dora. Voices starting with ef_ or em_ (ef_dora, em_alex, em_santa) speak Latin American Spanish; af_bella speaks English.
LanguageThe language the text is written in. Default: Spanish (es). The component doesn’t detect it, so pick the right one. It offers Spanish (also from Spain and Mexico), English (US and UK), French, Italian, Portuguese (also Brazilian), German, Japanese, Korean and Chinese.

Advanced options

FieldWhat it does
SpeedThe reading speed, from 0.5 to 2. Default: 1, the voice’s natural pace.

Output

The Data output returns a WAV audio file with the text read out. Its name is the first ten characters of the text.

Connect it to a component that stores or sends files. An agent can also use the component as a tool to create audio when it needs to; see what Tool Mode is in Key concepts.

Example

Every morning your flow summarises the open tickets with an agent. You connect the summary to Text to Speech, choose the voice em_alex and the language Spanish (es), and send the audio to the support team’s channel.

Good practices

  • Choose a voice that speaks the language of the text: an English voice reading Spanish text sounds forced.