跳转到主要内容
ElevenLabs brings world-class voice AI to ComfyUI, covering text-to-speech, speech-to-text, sound effects, voice isolation, and voice transformation. Its Eleven v3 and Multilingual v2 models deliver emotionally expressive speech across 29 languages, while Scribe v2 provides fast, accurate transcription with speaker diarization and entity detection.

Key capabilities

  • Text to Speech — Generate lifelike audio from text with three model tiers (Eleven v3, Multilingual v2, Flash v2.5)
  • Speech to Speech — Transform a source audio clip into a different voice while keeping the original intonation and pacing
  • Speech to Text — Transcribe audio with Scribe v2: 90+ languages, 32-speaker diarization, entity detection, keyword boost
  • Sound Effects — Generate audio effects from text descriptions
  • Voice Isolation — Separate voice from background noise, music, or ambient sound
  • Multi-Character Dialogue — Generate conversations between multiple distinct voices in a single pass
使用 API 节点需要保证你已经正常登录,并在受许可的网络环境下使用,请参考API 节点总览部分文档来了解使用 API 节点的具体使用要求。
请确保你的 ComfyUI 已经更新。本指南里的工作流可以在工作流模板中找到。如果找不到,可能是 ComfyUI 没有更新。如果加载工作流时有节点缺失,可能原因有:
  1. 你用的不是最新版(每夜版)。
  2. 启动时有些节点导入失败。

Models

ElevenLabs offers a suite of specialized voice AI models, each optimized for a distinct task. From high-fidelity speech generation to real-time transcription and audio editing, the following models are available as workflow-powered nodes in ComfyUI.

Eleven Multilingual v2 / v3

High-fidelity text-to-speech with three model tiers.

Scribe v2

Fast, accurate speech-to-text with speaker diarization.

Speech to Speech

Transform audio into a different voice while preserving delivery.

Text to Sound Effects

Generate sound effects and ambient audio from text.

Text to Dialogue

Multi-character dialogue with distinct voices in one pass.

Voice Isolation

Separate voice from background noise and music.