Skip to main content

Assistant voice

Voice characters

MapGPT's sophisticated voice synthesizers can breathe life into a plethora of characters. Be it replicating the tonal nuances of renowned celebrities or creating unique personalized voices tailored to your application's personality, the possibilities are boundless. Not limited to a monotonous robotic tone, our system is adept at delivering voices imbued with real human emotion and variation.

0:00
0:00

Hear for yourself. The sample above showcases a synthesized voice that's reminiscent of real-life speech.

Languages and Accents

MapGPT celebrates the diversity of its users by offering an extensive array of languages and accents:

  • English Variants: American (General & Southern), Australian, British (General & Essex), Indian, Irish, Italian English, and Scandinavian English.
  • World Languages: From Japanese, Chinese, and Hindi to Spanish, French, and Arabic, MapGPT supports a vast range of global languages.
  • Regional Touch: Beyond primary languages, the voice assistant is also able to capture regional dialects and nuances. Whether it's Swedish, Dutch, Finnish, or Korean, the tone is authentic and relatable.
0:00
0:00

Above is a sample of our language support, illustrating the natural tonality and fluency in Korean.

Customizing Voice with DashVoicesConfig

DashVoicesConfig allows you to finely tune the text-to-speech (TTS) systems used in your application. This feature provides the flexibility to fine tune the voices for your application's requirements.

Here's an example of how to use DashVoicesConfig to customize the voice settings:

Dash.init(
context = this,
accessToken = getString(R.string.mapbox_access_token),
) {
voices {
selectedVoice = PrebuiltDashVoices.voice1
preferLocalTts = true
localTtsEngine = "com.samsung.SMT"
}
}

Parameters

  • SelectedVoice: Specifies the remote voice for speech synthesis. Ignored when preferLocalTts is true.
  • PreferLocalTts: Chooses between local and remote voices. If true, only the local voice is used. If false, the SDK primarily uses the remote voice, switching to local in poor network conditions.
  • LocalTtsEngine: Defines the local TTS engine. Defaults to the system's default if the specified engine isn't found. Use Android's TextToSpeech to find available engines.

Runtime Updates

DashVoicesConfig supports updates at runtime, allowing the voice settings to be adjusted while the application is running.

Dash.applyUpdate {
voices {
preferLocalTts = settings.isLocalTtsEnabled
}
}
Was this page helpful?