Skip to content
AMC WebUI

Speech Synthesis & Transcription

Speech Synthesis & Transcription

AMC WebUI features expressively rich voice generation and speech-to-text recognition.

1. Text-to-Speech (TTS)

  • Default Model: gemini-3.1-flash-tts-preview.
  • 30 Character Voices:
    • Broad palette of expressive timbres, cadences, and styles (including Zephyr, Aoede, Charon, Fenrir, Kore, Puck, and Orpheus).
    • Preview voice samples and configure your default voice under Settings -> Language & Voice.
  • Message Readout: Click the speaker icon on any message bubble to synthesize and stream natural voice audio with an interactive media player.

2. Speech-to-Text Transcription

  • Default Model: gemini-3.5-transcribe.
  • Voice Typing: Click the microphone icon to dictate messages. Audio is converted into structured text with high accuracy across technical jargon and multi-lingual accents.