elevenlabs-stt

ElevenLabs scribe_v1 speech-to-text with auto language detection and optional speaker diarization. Suitable for subtitles, transcription, and meeting notes.

Overview

Field	Value
Model ID	`elevenlabs-stt`
CLI	`dlazy elevenlabs-stt`
MCP	`elevenlabs-stt` (Claude Code surfaces this as `mcp__dlazy__elevenlabs-stt`)
Type	`text`
Execution	Async task; the CLI polls until completion (`--no-wait` returns `generateId` immediately)
Batch	Supports `--batch <n>` parallel fan-out

Parameters

Arg	Type	Required	Notes
`audio_url`	url	Yes	Audio URL; audio (accepts URL, local path, or data: URL)
`language_code`	"zh" \| "en"	No	Language; default `"zh"`
`diarize`	string	No	Diarize Speakers; default `false`

--input @file.json or --input '{...}' can supply all args at once; flags take precedence over --input keys.

CLI Examples

dlazy elevenlabs-stt --help
dlazy elevenlabs-stt --audio_url "https://example.com/source.mp3"
dlazy elevenlabs-stt --audio_url "./local-audio.mp3"
dlazy elevenlabs-stt --audio_url "https://example.com/source.mp3" --dry-run
dlazy elevenlabs-stt --audio_url "https://example.com/source.mp3" --no-wait
dlazy elevenlabs-stt --audio_url "https://example.com/source.mp3" --batch 4

Compose with a pipeline

dlazy gemini-2.5-tts --prompt "reference voice line" \
  | dlazy elevenlabs-stt --audio_url -

MCP is consumed by AI clients, not handwritten. Once dLazy is registered as an MCP server the tool appears in the client's tool list — Claude Code surfaces it as mcp__dlazy__elevenlabs-stt; generic clients (e.g. OpenClaw) call it as elevenlabs-stt.

See MCP setup for how to add the server.

Output

{
  "outputs": [
    { "type": "json", "id": "o_...", "value": { /* tool-specific payload */ } }
  ]
}

Structured payloads are returned as json outputs; the inner shape is tool-specific.