elevenlabs-dialogue

name	elevenlabs-dialogue
description	Create multi-speaker dialogue audio. Use for: podcasts, conversations, audiobook scenes
allowed-tools	Bash(curl *)
disable-model-invocation	true

Generate multi-speaker dialogue audio using ElevenLabs text-to-dialogue endpoint. Each segment can use a different voice. Up to 10 unique voices per request.

Ask the user for:

• SEGMENTS: list of dialogue lines, each with text and speaker (e.g. "Alice: Hello! / Bob: Hi there!")
• OUTPUT_FILE (optional): where to save the MP3 — default: ~/Downloads/dialogue_[timestamp].mp3

Available voices:

Name	Voice ID	Accent	Gender
George (default male)	JBFqnCBsd6RMkjVDRZzb	British	Male
Rachel	21m00Tcm4TlvDq8ikWAM	American	Female
Aria	9BWtsMINqrJLrRacOk9x	American	Female
Sarah	EXAVITQu4vr4xnSDxMaL	American	Female
Charlie	IKne3meq5aSn9XLyUdCD	Australian	Male

Then run the following steps:

Step 1 — Check API key:

(bash) if [ -z "$ELEVENLABS_API_KEY" ]; then echo "ERROR: ELEVENLABS_API_KEY not set." echo "Run /pocket-knife:setup to configure your API keys." exit 1 fi

OUTPUT_FILE="${OUTPUT_FILE:-$HOME/Downloads/dialogue_$(date +%Y%m%d_%H%M%S).mp3}" (end bash)

Step 2 — Generate dialogue audio:

(bash)

Source: https://elevenlabs.io/docs/api-reference/text-to-dialogue/convert

Use -f (NOT --fail-with-body) for binary output — error body would corrupt the MP3 file

Model: eleven_v3 (NOT eleven_monolingual_v1 — deprecated)

inputs array: max 10 unique voice_ids per request

Replace the inputs array below with the actual dialogue segments from the user

curl -f -s
-X POST "https://api.elevenlabs.io/v1/text-to-dialogue"
-H "xi-api-key: $ELEVENLABS_API_KEY"
-H "Content-Type: application/json"
-d '{ "inputs": [ {"text": "[FIRST_LINE_TEXT]", "voice_id": "JBFqnCBsd6RMkjVDRZzb"}, {"text": "[SECOND_LINE_TEXT]", "voice_id": "EXAVITQu4vr4xnSDxMaL"} ], "model_id": "eleven_v3" }'
--output "$OUTPUT_FILE"

if [ $? -ne 0 ]; then echo "ERROR: ElevenLabs Dialogue API call failed." echo "Check that ELEVENLABS_API_KEY is correct in ~/.claude/.env" echo "Run /pocket-knife:setup to reconfigure." exit 1 fi

Verify the output file has content

if [ ! -s "$OUTPUT_FILE" ]; then echo "ERROR: Audio file is empty. Possible causes:" echo " - Invalid voice_id in inputs array" echo " - API returned error without proper HTTP status code" rm -f "$OUTPUT_FILE" exit 1 fi

echo "Dialogue audio generated successfully: $OUTPUT_FILE" (end bash)

Report to the user:

• The path where the audio file was saved
• Number of dialogue segments generated
• Voices used per segment
• If any error: missing key run /pocket-knife:setup; empty file check voice IDs