Guides
Voice Configuration

Voice Configuration

Configure text-to-speech voices for natural avatar speech.

TTS Providers

ProviderVoicesQualityLatency
Google (default)50+HighFast
ElevenLabs100+PremiumMedium
BrowserSystemVariableFastest

Configuration

<Avatar
  apiKey="..."
  avatarId="aria-001"
  ttsProvider="elevenlabs"
  voiceId="21m00Tcm4TlvDq8ikWAM"
/>

Google Cloud TTS

Google is the default TTS provider, offering a good balance of quality and speed.

Popular Voices

Voice IDNameLanguage
en-US-Neural2-FFemale US Englishen-US
en-US-Neural2-DMale US Englishen-US
en-GB-Neural2-AFemale Britishen-GB
en-AU-Neural2-AFemale Australianen-AU
<Avatar
  ttsProvider="google"
  voiceId="en-US-Neural2-F"
/>

ElevenLabs

ElevenLabs offers the most natural-sounding voices.

Popular Voices

Voice IDNameStyle
21m00Tcm4TlvDq8ikWAMRachelWarm, conversational
EXAVITQu4vr4xnSDxMaLBellaSoft, friendly
ErXwobaYiN019PkySvjVAntoniProfessional, clear
<Avatar
  ttsProvider="elevenlabs"
  voiceId="21m00Tcm4TlvDq8ikWAM"
/>
⚠️

ElevenLabs voices are only available on Creator and Pro plans.

Browser TTS

Use the browser's built-in speech synthesis:

<Avatar
  ttsProvider="browser"
/>

Browser TTS is:

  • Free (doesn't count against quota)
  • Fastest (no network latency)
  • Variable quality (depends on OS/browser)

Voice Settings

Fine-tune voice parameters:

<Avatar
  ttsProvider="google"
  voiceId="en-US-Neural2-F"
  voiceSettings={{
    pitch: 0,      // -20 to 20
    speakingRate: 1.0,  // 0.25 to 4.0
  }}
/>

Language Detection

Auto-detect language and select appropriate voice:

<Avatar
  autoLanguage={true}
  defaultVoice="en-US-Neural2-F"
  languageVoices={{
    en: 'en-US-Neural2-F',
    es: 'es-ES-Neural2-A',
    fr: 'fr-FR-Neural2-A',
    de: 'de-DE-Neural2-A',
  }}
/>