ElevenLabs

Remove background noise from audio, isolating speech using ElevenLabs. Upload noisy audio and receive a clean version with background noise removed. Useful for improving recording quality, cleaning up meeting recordings, or preparing audio for further processing. Returns a durable `asset_id` and an authenticated in-app URL to the cleaned audio file. To deliver it to a person, pass `asset_id` via the `attachments` parameter of respond/send_to; do not paste the URL into message body text.

Translate audio/video content into other languages while preserving speaker voice and emotion using ElevenLabs. This tool has two modes. Supply either the first pair or dubbing_id, never neither: 1. Submit a new dubbing job: provide file_url and target_language (dubbing_id omitted). Returns a dubbing_id. 2. Check dubbing status: provide dubbing_id alone to poll for completion. When complete, returns a durable asset_id and an authenticated in-app URL to the dubbed file. To deliver it to a person, pass asset_id via the attachments parameter of respond/send_to; do not paste the URL into message body text. Supports 32+ languages. Max file size: 50MB, 2.5 hours.

Generate music tracks from text descriptions using ElevenLabs. Create background music, compositions, and tracks by describing the desired style, mood, and instruments. Supports duration control (5-300 seconds) and instrumental-only mode. Returns a durable `asset_id` and an authenticated in-app URL to the generated audio file. To deliver it to a person, pass `asset_id` via the `attachments` parameter of respond/send_to; do not paste the URL into message body text.

Generate sound effects from text descriptions using ElevenLabs. Create any sound effect by describing it in natural language (e.g., 'thunder rumbling in the distance', 'gentle notification chime', 'keyboard typing sounds'). Supports duration control (0.5-30 seconds) and looping. Returns a durable `asset_id` and an authenticated in-app URL to the generated audio file. To deliver it to a person, pass `asset_id` via the `attachments` parameter of respond/send_to; do not paste the URL into message body text.

Convert text to natural-sounding speech audio using ElevenLabs. Accepts a connected-account voice name or direct voice ID; if omitted, selects the first available account voice. Unknown names return the available account voices. Supports 4 TTS models (flash for speed, turbo for balance, multilingual_v2 for quality, v3 for emotion) and 3 output formats (mp3, pcm, opus). Returns a durable `asset_id` and an authenticated in-app URL to the generated audio file. To deliver it to a person, pass `asset_id` via the `attachments` parameter of respond/send_to; do not paste the URL into message body text.

Convert audio or video files to text transcripts using ElevenLabs. Supports speaker diarization, word/character-level timestamps, and multiple output formats (text, srt subtitles, json with timestamps). Accepts MP3, WAV, M4A, MP4, WebM, and other common audio/video formats up to 50MB.

Transform the voice in an audio recording to a different voice while preserving speech content using ElevenLabs. Upload audio and optionally select a connected-account voice name or direct voice ID. If omitted, the first available account voice is selected; unknown names return the available account voices. Returns a durable `asset_id` and an authenticated in-app URL to the transformed audio file. To deliver it to a person, pass `asset_id` via the `attachments` parameter of respond/send_to; do not paste the URL into message body text.