Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/phpaisdk/elevenlabs/llms.txt

Use this file to discover all available pages before exploring further.

The aisdk/elevenlabs package deliberately targets ElevenLabs’ generative and direct media APIs — the surfaces where you create, transform, or analyse audio content. Administrative, library-management, and platform-infrastructure APIs are intentionally left out. This keeps the package focused on the operations most applications actually need, and avoids coupling your code to ElevenLabs account-management workflows that have nothing to do with media generation.

Included in this package

TTS, batch STT, realtime Scribe, voice transforms, music, sound effects, dialogue, dubbing, and forced alignment — the full generative media surface.

Excluded APIs

Administrative, library, and platform APIs are out of scope. Use the official ElevenLabs PHP SDK or an HTTP client directly for those needs.

​
What Is Included

The following capabilities are fully supported by this package:
  • Text-to-speech — generate speech from text using any ElevenLabs TTS model via Generate::speech() and ElevenLabs::model().
  • Batch transcription (Scribe v2) — transcribe audio files via Generate::transcription().
  • Realtime transcription — stream audio to Scribe v2 Realtime using the core Live::transcribe() API with this package’s transport encoding.
  • Voice changer — convert audio to a different voice via ElevenLabs::voiceChanger()->convert().
  • Voice isolation — strip background noise and isolate speech via ElevenLabs::voiceIsolator()->isolate().
  • Voice design — generate voice previews from a text description via ElevenLabs::voiceDesign()->design().
  • Voice remixing — transform an eligible existing voice into new design variants via ElevenLabs::voiceDesign()->remix().
  • Saving a generated voice preview — persist a preview as a permanent usable voice via ElevenLabs::voiceDesign()->create(). This is included because it is the required second step of the Voice Design and Remix APIs — without it, neither feature is useful.
  • Music generation — compose music from a text prompt, generate from a plan, or produce detailed output with composition metadata via ElevenLabs::music().
  • Direct music transforms — convert video to a music soundtrack (videoToMusic), upload source audio for composition-plan workflows (upload), and separate audio into stems (separateStems).
  • Sound effects — generate sound effects from a text description via ElevenLabs::soundEffects()->generate().
  • Text-to-dialogue — synthesise multi-voice dialogue scenes via ElevenLabs::dialogue()->create() or withTimestamps().
  • Dubbing creation and retrieval — submit a dubbing job, poll its status, and retrieve dubbed video or transcripts via ElevenLabs::dubbing().
  • Forced alignment — align an audio file against a known transcript at character and word level via ElevenLabs::forcedAlignment()->create().

​
What Is Deliberately Excluded

The following ElevenLabs API surfaces are outside the scope of this package:
  • ElevenAgents — the conversational AI agent platform and its associated telephony, runtime configuration, and knowledge-base management APIs.
  • Voice library — browsing the shared voice marketplace, cloning voices from samples, managing voice training jobs, and general voice CRUD (listing, updating, or deleting voices).
  • Music asset management — music finetuning, history and asset management, and marketplace-publishing workflows.
  • Studio, Audio Native, Flows — ElevenLabs project and editor management, Audio Native embed configuration, and Flows pipeline management.
  • Dubbing Studio editing — resource-level editing, listing all dubbing projects, and deleting dubbing jobs. Only job creation, status polling, and output retrieval are included.
  • Account and workspace administration — user account settings, workspace configuration, API-key management, usage analytics, billing, and administrative endpoints.
  • Pronunciation-dictionary management — creating, updating, or applying custom pronunciation dictionaries.
  • AI-audio detection — the ElevenLabs AI speech detection classifier.
  • Human production services — the ElevenLabs professional voice cloning and human production ordering APIs.
  • HTTP streaming variants — streaming chunked-transfer variants of otherwise one-shot media endpoints (e.g. streamed TTS) are not included; use Generate::speech() for standard TTS output.
Keeping the package boundary at the generative media surface prevents scope creep that would require versioning this package every time ElevenLabs changes an admin API unrelated to media generation. Applications that need excluded APIs should use the official ElevenLabs PHP SDK or a PSR-18 HTTP client pointed at the ElevenLabs API directly.

Build docs developers (and LLMs) love

Get started for freeTalk to us