# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Elevenlabs Music Generation API
  version: 1.0.0
extends: openapi/elevenlabs-music-generation-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-09-26'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 7
- target: $.paths['/v1/music'].post
  update:
    x-apievangelist-phrasing:
      intent: Generate a song from a text prompt
      effect: write
      questions:
      - Can ElevenLabs make an original song from a description of the mood and genre?
      - Is it possible to generate an instrumental track that loops?
      - How long can a generated music track be?
      instructions:
      - text: Generate music from the prompt {prompt}.
        slots:
          prompt: requestBody.prompt
      - text: 'Make a {duration_seconds}-second instrumental track: {prompt}.'
        slots:
          duration_seconds: requestBody.duration_seconds
          prompt: requestBody.prompt
      - text: Create a loopable piece of music described as {prompt}, loop {loop}.
        slots:
          prompt: requestBody.prompt
          loop: requestBody.loop
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/music/plan'].post
  update:
    x-apievangelist-phrasing:
      intent: Draft a composition plan for a song
      effect: read
      questions:
      - Can I get a section-by-section plan for a song before generating audio?
      - Is there a way to revise an existing composition plan with a new prompt?
      instructions:
      - text: Generate a composition plan for {prompt}.
        slots:
          prompt: requestBody.prompt
      - text: Plan a {music_length_ms} ms song from prompt {prompt}.
        slots:
          music_length_ms: requestBody.music_length_ms
          prompt: requestBody.prompt
      - text: Revise composition plan {source_composition_plan} using prompt {prompt}.
        slots:
          source_composition_plan: requestBody.source_composition_plan
          prompt: requestBody.prompt
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/music/detailed'].post
  update:
    x-apievangelist-phrasing:
      intent: Compose a song with timestamps and metadata
      effect: write
      questions:
      - Can I get word timestamps and a waveform along with a composed song?
      - How do I compose a song from my own lyrics and a composition plan?
      instructions:
      - text: Compose a song from plan {composition_plan} and return detailed metadata.
        slots:
          composition_plan: requestBody.composition_plan
      - text: Compose {prompt} with lyric timestamps {with_timestamps} and a waveform visual {with_waveform_visual}.
        slots:
          prompt: requestBody.prompt
          with_timestamps: requestBody.with_timestamps
          with_waveform_visual: requestBody.with_waveform_visual
      - text: Compose a song with lyrics {lyrics_text} in the style {music_prompt}, signed with C2PA {sign_with_c2pa}.
        slots:
          lyrics_text: requestBody.lyrics_text
          music_prompt: requestBody.music_prompt
          sign_with_c2pa: requestBody.sign_with_c2pa
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/music/detailed/stream'].post
  update:
    x-apievangelist-phrasing:
      intent: Stream a song with metadata over SSE
      effect: write
      questions:
      - Can I receive a song and its timestamps as server-sent events while it generates?
      - Which music call streams audio together with detailed metadata?
      instructions:
      - text: Stream a detailed composition of {prompt} as server-sent events.
        slots:
          prompt: requestBody.prompt
      - text: Stream composition plan {composition_plan} over SSE with timestamps {with_timestamps}.
        slots:
          composition_plan: requestBody.composition_plan
          with_timestamps: requestBody.with_timestamps
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/music/stream'].post
  update:
    x-apievangelist-phrasing:
      intent: Stream composed music audio
      effect: write
      questions:
      - How do I start playing a generated song before it finishes composing?
      - Can I stream just the audio of a song made from a plan, without extra metadata?
      instructions:
      - text: Stream the audio of a song composed from {prompt}.
        slots:
          prompt: requestBody.prompt
      - text: Stream audio only for plan {composition_plan} in output format {output_format}.
        slots:
          composition_plan: requestBody.composition_plan
          output_format: query.output_format
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/music/upload'].post
  update:
    x-apievangelist-phrasing:
      intent: Upload a song for later inpainting
      effect: write
      questions:
      - Can I upload my own track so I can edit parts of it later?
      - Is uploading a song charged, and is it checked for copyright?
      instructions:
      - text: Upload music file {file} for inpainting.
        slots:
          file: requestBody.file
      - text: Upload {file} and extract its composition plan ({extract_composition_plan}).
        slots:
          file: requestBody.file
          extract_composition_plan: requestBody.extract_composition_plan
      method: generated
      generated: '2026-09-26'
- target: $.paths['/v1/music/stem-separation'].post
  update:
    x-apievangelist-phrasing:
      intent: Split a song into separate stems
      effect: write
      questions:
      - Can I separate vocals, drums and other instruments from a track?
      - Why does stem separation take a long time on long songs?
      instructions:
      - text: Split {file} into stems.
        slots:
          file: requestBody.file
      - text: Separate {file} into stems using variation {stem_variation_id}.
        slots:
          file: requestBody.file
          stem_variation_id: requestBody.stem_variation_id
      method: generated
      generated: '2026-09-26'