Deepgram Flux TTS (the /v2/speak endpoint).

Sits alongside TTS, the Aura client for /v1/speak; this class does not replace it. Both the streaming (TTSv2.stream) and batch (TTSv2.synthesize) paths emit linear16 only — Deepgram also offers mp3/opus/flac/aac, but the LiveKit pipeline plays raw PCM and this plugin ships no decoder, so any other encoding is rejected.

Example

import { TTSv2 } from '@livekit/agents-plugin-deepgram';

const tts = new TTSv2({ model: 'flux-alexis-en' });

Hierarchy (view full)

Constructors

Properties

label: string = 'deepgram.TTSv2'

Accessors

  • get capabilities(): TTSCapabilities
  • Returns this TTS's capabilities

    Returns TTSCapabilities

  • get markup(): TTSMarkup
  • Access TTS markup capabilities (instructions for the LLM, text conversion).

    Returns TTSMarkup

  • get model(): string
  • Get the model name/identifier for this TTS instance.

    Returns string

    The model name if available, "unknown" otherwise.

    Remarks

    Plugins should override this property to provide their model information.

  • get numChannels(): number
  • Returns the channel count of audio frames returned by this TTS

    Returns number

  • get provider(): string
  • Get the provider name for this TTS instance.

    Returns string

    The provider name if available, "unknown" otherwise.

    Remarks

    Plugins should override this property to provide their provider information.

  • get sampleRate(): number
  • Returns the sample rate of audio frames returned by this TTS

    Returns number

Methods

  • Open the pooled WebSocket ahead of the first generation so the first turn does not pay the connect. Safe to call more than once.

    Returns void