Provider Elevenlabs

ElevenLabs

Mode: audio · Models: 10

Vendor: elevenlabs.io · Official API docs: API reference

ElevenLabs is the most complete voice suite in the catalog: expressive text-to-speech, multilingual narration, sound effects, voice design, dubbing, speech-to-speech conversion, and audio isolation.

Models

id	Name	Input type	Use
eleven-v3	Eleven v3	tts	Most expressive TTS
eleven-multilingual-v2	Eleven Multilingual v2	tts	Many languages
eleven-voice-design-v3	Eleven Voice Design v3	tts	Design a new voice
eleven-voice-design-multilingual-v2	Voice Design Multilingual v2	tts	Multilingual voice design
elevenlabs-sfx	ElevenLabs SFX v2	sfx	Sound effects
eleven-sts-v2	Eleven STS v2	sts	Speech-to-speech
eleven-multilingual-sts-v2	Eleven Multilingual STS v2	sts	Multilingual STS
eleven-dubbing	Eleven Dubbing	sts	Dub a track
eleven-audio-isolation	Eleven Audio Isolation	sts	Isolate voice from noise
eleven-voice-previews	Eleven Voice Previews	tts	Preview voices

CLI

# text-to-speech
gen-ai generate -m eleven-v3 -p "Welcome to Picsart AI Playground." --voice JBFqnCBsd6RMkjVDRZzb

# sound effect from a description
gen-ai generate -m elevenlabs-sfx -p "a heavy wooden door creaking open"

MCP

{ "name": "picsart_generate",
  "arguments": {
    "model": "eleven-v3",
    "prompt": "Welcome to Picsart AI Playground.",
    "extra": { "voiceId": "JBFqnCBsd6RMkjVDRZzb", "language": "en" }
  } }

Parameters — eleven-v3

Param	CLI flag	Type	Notes
prompt	-p	text	required — the text to speak
voiceId	--voice	enum	20 preset voice ids (see below)
language	--language	text	language / locale
accent	--accent	text	accent hint

List the full, current voice catalog any time:

gen-ai models info eleven-v3 --json | jq '.paramConfig.voiceId.descriptor.options'

Source: gen-ai models info eleven-v3 --json. Voice ids are 20 presets such as JBFqnCBsd6RMkjVDRZzb, EkK5I93UQWFDigLMpZcX, RILOU7YmBhvwJGDGjNmP — verify the live list before hard-coding.

Pricing

gen-ai pricing eleven-v3 -p "your script here"

Cost scales with the length of the generated audio.