ElevenLabs Eleven v4 API, ElevenLabs Music API, ElevenLabs Sound Effects, Dubbing, Voice Changer, Multilingual v2 API: Pricing, Documentation

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-eleven-v4-tts/v1/text-to-speech' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"text": "Welcome to Pixazo. [whispers] This is Eleven v4, the most expressive voice model from ElevenLabs.", "stability": 0.5}'

Callback Payload (success)

{
  "request_id": "elevenlabs-eleven-v4-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-eleven-v4-tts",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-30T10:04:54.387Z",
  "completed_at": "2026-09-30 10:05:14"
}

Failure callback shape

{
  "request_id": "elevenlabs-eleven-v4-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-eleven-v4-tts",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll while the request is processing (status PROCESSING), plus the final COMPLETED/ERROR callback.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

• text is the only required field, 1–5000 characters. Spaces, punctuation and audio tags count towards the length; an emoji counts as two characters.
• Omit optional fields you do not use — do not send them as null.
• Eleven v4 has two voice settings, stability and similarity_boost. There is no style, speed or speaker-boost setting, and SSML tags such as <break> are not supported — use audio tags such as [pause] instead.
• timestamps (per-character timing) is not available on this endpoint. A request with timestamps: true is rejected.
• ElevenLabs keeps improving Eleven v4 after launch, so the same text can sound slightly different over time.

Audio tags

Write a short instruction in square brackets where the delivery should change — an emotion such as [excited], a way of speaking such as [whispers], a reaction such as [laughs], or a pause such as [pause]. Tags are free-form: describe the voice you want in plain words. Because the model can also produce sound effects, a tag that names a sound may be read as one, so describe how the voice should sound.

{
  "text": "[excited] We did it! [pause] [whispers] Now keep it a secret.",
  "language_code": "en"
}

Request Headers

Response Handling

Most parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. A few are only known once the provider has the request; they are reported through the status endpoint as status: "ERROR", with a message in error saying what to fix. Failed requests are not billed — the hold is released in full.

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED or ERROR, then download the speech from output.media_url. Treat any status other than QUEUED, PROCESSING or COMPLETED as a failure.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-eleven-v4-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-eleven-v4-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-eleven-v4-tts",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-30T10:04:54.387Z",
  "completed_at": "2026-09-30 10:05:14"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

How billing is measured

Billing is measured on the length of the text you submit: every started block of 1,000 characters — spaces, punctuation and audio tags included — is one billing unit, so a short request bills one block. The amount is placed as a hold when the request is submitted and released in full if the request fails.

The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-eleven-v4-turbo-tts/v1/text-to-speech' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"text": "Thanks for calling Pixazo. [warmly] How can I help you today?", "stability": 0.5}'

Callback Payload (success)

{
  "request_id": "elevenlabs-eleven-v4-turbo-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-eleven-v4-turbo-tts",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-30T10:04:54.387Z",
  "completed_at": "2026-09-30 10:05:14"
}

Failure callback shape

{
  "request_id": "elevenlabs-eleven-v4-turbo-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-eleven-v4-turbo-tts",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll while the request is processing (status PROCESSING), plus the final COMPLETED/ERROR callback.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

• text is the only required field, 1–5000 characters. Spaces, punctuation and audio tags count towards the length; an emoji counts as two characters.
• Omit optional fields you do not use — do not send them as null.
• Eleven v4 Turbo has two voice settings, stability and similarity_boost. There is no style, speed or speaker-boost setting, and SSML tags such as <break> are not supported — use audio tags such as [pause] instead.
• timestamps (per-character timing) is not available on this endpoint. A request with timestamps: true is rejected.
• ElevenLabs keeps improving Eleven v4 after launch, so the same text can sound slightly different over time.

Audio tags

Write a short instruction in square brackets where the delivery should change — an emotion such as [excited], a way of speaking such as [whispers], a reaction such as [laughs], or a pause such as [pause]. Tags are free-form: describe the voice you want in plain words. Because the model can also produce sound effects, a tag that names a sound may be read as one, so describe how the voice should sound.

{
  "text": "[excited] We did it! [pause] [whispers] Now keep it a secret.",
  "language_code": "en"
}

Request Headers

Response Handling

Most parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. A few are only known once the provider has the request; they are reported through the status endpoint as status: "ERROR", with a message in error saying what to fix. Failed requests are not billed — the hold is released in full.

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED or ERROR, then download the speech from output.media_url. Treat any status other than QUEUED, PROCESSING or COMPLETED as a failure.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-eleven-v4-turbo-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-eleven-v4-turbo-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-eleven-v4-turbo-tts",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-30T10:04:54.387Z",
  "completed_at": "2026-09-30 10:05:14"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

How billing is measured

Billing is measured on the length of the text you submit: every started block of 1,000 characters — spaces, punctuation and audio tags included — is one billing unit, so a short request bills one block. The amount is placed as a hold when the request is submitted and released in full if the request fails.

The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-music/v2.5/text-to-music' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"prompt": "Upbeat lofi hip hop with warm piano chords and a relaxed drum groove", "music_length_ms": 30000}'

Callback Payload (success)

{
  "request_id": "elevenlabs-music-v2-5_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-music-v2-5",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Failure callback shape

{
  "request_id": "elevenlabs-music-v2-5_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-music-v2-5",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on each status change.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Prompt or composition plan

• Send exactly one of prompt or composition_plan. Sending both, or neither, fails the request.
• music_length_ms and force_instrumental apply to prompt only. With a composition plan, the length is the sum of the chunk durations.
• The model is fixed by the URL. Do not send a model_id in the body — it is rejected. To use a different version, call its URL.
• The track is at most 10 minutes long and is returned as an MP3.

Composition plans

A composition plan describes the track as an ordered list of chunks, each with its own lyrics, length and style. Use it when you need control over structure — an intro, verses, a chorus — rather than leaving the arrangement to the model.

Example

{
  "composition_plan": {
    "chunks": [
      {
        "text": "[Intro]",
        "duration_ms": 8000,
        "positive_styles": [
          "cinematic",
          "soft strings",
          "slow build"
        ]
      },
      {
        "text": "[Verse 1]\nNeon rain on empty streets\nI keep walking to the beat",
        "duration_ms": 20000,
        "positive_styles": [
          "synthwave",
          "analog drum machine",
          "male vocals"
        ],
        "negative_styles": [
          "distorted guitar"
        ]
      }
    ]
  }
}

Migrating from ElevenLabs Music v1

The v1 composition plan used a sections array. This version uses chunks, and the two shapes are not interchangeable: a plan written for v1 fails here, and a plan written for this version fails on v1. Prompt-only requests work the same way on both.

Request Headers

Response Handling

Most parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. A few checks need the whole composition plan to be read, so they run after the request is accepted and are reported through the status endpoint as status: "ERROR", with a message in error saying what to fix. Failed requests are not billed — the hold is released in full.

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED (or FAILED/ERROR), then download the track from output.media_url.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-music-v2-5_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-music-v2-5_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-music-v2-5",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or FAILED/ERROR (failure).

How billing is measured

Billing is measured on the length of the generated track, rounded up to the next whole minute — not on how long generation takes. A hold is placed when the request is submitted, sized from music_length_ms or from the chunk durations of a composition plan; if neither is given, the hold covers the 10-minute maximum. The hold is reduced to the real cost once the finished track has been measured, and released in full if the request fails.

The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-music/v2/text-to-music' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"prompt": "Upbeat lofi hip hop with warm piano chords and a relaxed drum groove", "music_length_ms": 30000}'

Callback Payload (success)

{
  "request_id": "elevenlabs-music-v2_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-music-v2",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Failure callback shape

{
  "request_id": "elevenlabs-music-v2_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-music-v2",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on each status change.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Prompt or composition plan

• Send exactly one of prompt or composition_plan. Sending both, or neither, fails the request.
• music_length_ms and force_instrumental apply to prompt only. With a composition plan, the length is the sum of the chunk durations.
• The model is fixed by the URL. Do not send a model_id in the body — it is rejected. To use a different version, call its URL.
• The track is at most 10 minutes long and is returned as an MP3.

Composition plans

A composition plan describes the track as an ordered list of chunks, each with its own lyrics, length and style. Use it when you need control over structure — an intro, verses, a chorus — rather than leaving the arrangement to the model.

Example

{
  "composition_plan": {
    "chunks": [
      {
        "text": "[Intro]",
        "duration_ms": 8000,
        "positive_styles": [
          "cinematic",
          "soft strings",
          "slow build"
        ]
      },
      {
        "text": "[Verse 1]\nNeon rain on empty streets\nI keep walking to the beat",
        "duration_ms": 20000,
        "positive_styles": [
          "synthwave",
          "analog drum machine",
          "male vocals"
        ],
        "negative_styles": [
          "distorted guitar"
        ]
      }
    ]
  }
}

Migrating from ElevenLabs Music v1

The v1 composition plan used a sections array. This version uses chunks, and the two shapes are not interchangeable: a plan written for v1 fails here, and a plan written for this version fails on v1. Prompt-only requests work the same way on both.

Request Headers

Response Handling

Most parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. A few checks need the whole composition plan to be read, so they run after the request is accepted and are reported through the status endpoint as status: "ERROR", with a message in error saying what to fix. Failed requests are not billed — the hold is released in full.

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED (or FAILED/ERROR), then download the track from output.media_url.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-music-v2_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-music-v2_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-music-v2",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or FAILED/ERROR (failure).

How billing is measured

Billing is measured on the length of the generated track, rounded up to the next whole minute — not on how long generation takes. A hold is placed when the request is submitted, sized from music_length_ms or from the chunk durations of a composition plan; if neither is given, the hold covers the 10-minute maximum. The hold is reduced to the real cost once the finished track has been measured, and released in full if the request fails.

The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-video-to-music/v1/video-to-music' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"video_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/doc-assets/video/input-17s.mp4", "description": "Warm, hopeful acoustic score that builds gently", "tags": ["cinematic", "acoustic"]}'

Callback Payload (success)

{
  "request_id": "elevenlabs-video-to-music_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-video-to-music",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Failure callback shape

{
  "request_id": "elevenlabs-video-to-music_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-video-to-music",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on each status change.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

• Send exactly one of video_url or video_urls.
• Videos must be MP4 or MOV, at most 600 seconds (10 minutes) and 32 MB in total, at public http(s) URLs that stay reachable until the request has been processed.
• This endpoint always uses ElevenLabs Music v2.5. Do not send a model_id in the body — it is rejected.

Input videos

• The videos are downloaded when the request is processed, so the URLs must stay reachable until then.
• Several videos in video_urls are combined in the order given, and one track is generated for the combined video.
• The length of MP4 and MOV videos is read when you submit, so videos over 600 seconds in total are rejected straight away and the hold placed on your balance matches the videos. If the length of a single video_url cannot be read, the hold covers the 10-minute maximum until the track has been measured.

Request Headers

Response Handling

Most parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. A few checks run after the request is accepted and are reported through the status endpoint as status: "ERROR", with a message in error saying what to fix. Failed requests are not billed — the hold is released in full.

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED (or FAILED/ERROR), then download the track from output.media_url.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-video-to-music_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-video-to-music_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-video-to-music",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or FAILED/ERROR (failure).

How billing is measured

Billing is measured on the length of the generated track, rounded up to the next whole minute — not on how long generation takes. A hold sized from the combined length of the videos is placed when the request is submitted; if the length of a single video_url cannot be read, the hold covers the 10-minute maximum. The hold is reduced to the real cost once the track has been measured, and released in full if the request fails.

The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-sound-effects/v1/text-to-sound-effects' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"text": "Heavy wooden door creaking open slowly in a large empty hall", "duration_seconds": 5}'

Callback Payload (success)

{
  "request_id": "elevenlabs-sound-effects_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-sound-effects",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Failure callback shape

{
  "request_id": "elevenlabs-sound-effects_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-sound-effects",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on each status change.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

• text is the only required field. The sound effect is at most 30 seconds long.
• The model is fixed by the URL. Do not send a model_id in the body — it is rejected.

Request Headers

Response Handling

Most parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. A few checks run after the request is accepted and are reported through the status endpoint as status: "ERROR", with a message in error saying what to fix. Failed requests are not billed — the hold is released in full.

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED (or FAILED/ERROR), then download the sound effect from output.media_url.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-sound-effects_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-sound-effects_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-sound-effects",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or FAILED/ERROR (failure).

How billing is measured

Billing is measured on the length of the generated sound effect, in seconds — not on how long generation takes. A hold covering the 30-second maximum is placed when the request is submitted; it is reduced to the real cost once the sound effect has been measured, and released in full if the request fails.

The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Add the X-Webhook-URL header to your generate request to receive a POST callback instead of polling.

X-Webhook-URL: https://your-server.com/webhook/callback

Request Parameters - ElevenLabs Music generate request

Request Headers

Response Handling

Common status codes.

Error Responses

Queue system errors and model validation errors.

Queue System Errors

// 402 — Insufficient balance
{
  "error": "Insufficient Balance",
  "message": "Your wallet does not have enough balance."
}

// 400 — Model not found
{
  "error": "Model not found",
  "message": "Model 'elevenlabs-music' not found or is disabled"
}

Error via Status/Webhook

{
  "request_id": "elevenlabs-music_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-music",
  "error": "Description of the error",
  "output": null
}

Retrieving Results

Poll the universal status endpoint to check progress and retrieve results.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_API_KEY

curl -H "Ocp-Apim-Subscription-Key: YOUR_API_KEY" \
  "https://gateway.pixazo.ai/v2/requests/status/elevenlabs-music_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

{
  "request_id": "elevenlabs-music_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-music",
  "error": null,
  "output": {
    "media_url": ["https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/elevenlabs-music_019dxxxx/output.mp3"],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-03-31T10:00:00.000Z",
  "updated_at": "2026-03-31T10:00:15.000Z",
  "completed_at": "2026-03-31T10:00:15.000Z"
}

Response Fields

Status Values

Status Flow

QUEUED → PROCESSING → COMPLETED
                    → FAILED
                    → ERROR

Typical Workflow

1. Send a generate request to the API endpoint
2. Save the request_id from the response
3. Poll every 5-10 seconds: GET /v2/requests/status/{request_id}
4. When status is "COMPLETED", download from output.media_url

Tip: Use X-Webhook-URL header to get a callback instead of polling.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Add the X-Webhook-URL header to your generate request to receive a POST callback instead of polling.

X-Webhook-URL: https://your-server.com/webhook/callback

Request Parameters - ElevenLabs Eleven v3 TTS generate request

Content Item Types & Limits

Request Headers

Response Handling

Common status codes.

Error Responses

Queue system errors and model validation errors.

Queue System Errors

// 402 — Insufficient balance
{
  "error": "Insufficient Balance",
  "message": "Your wallet does not have enough balance."
}

// 400 — Model not found
{
  "error": "Model not found",
  "message": "Model 'elevenlabs-eleven-v3-tts' not found or is disabled"
}

Error via Status/Webhook

{
  "request_id": "elevenlabs-eleven-v3-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-eleven-v3-tts",
  "error": "Description of the error",
  "output": null
}

Retrieving Results

Poll the universal status endpoint to check progress and retrieve results.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_API_KEY

curl -H "Ocp-Apim-Subscription-Key: YOUR_API_KEY" \
  "https://gateway.pixazo.ai/v2/requests/status/elevenlabs-eleven-v3-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

{
  "request_id": "elevenlabs-eleven-v3-tts_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-eleven-v3-tts",
  "error": null,
  "output": {
    "media_url": ["https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/elevenlabs-eleven-v3-tts_019dxxxx/output.mp3"],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-03-31T10:00:00.000Z",
  "updated_at": "2026-03-31T10:00:15.000Z",
  "completed_at": "2026-03-31T10:00:15.000Z"
}

Response Fields

Status Values

Status Flow

QUEUED → PROCESSING → COMPLETED
                    → FAILED
                    → ERROR

Typical Workflow

1. Send a generate request to the API endpoint
2. Save the request_id from the response
3. Poll every 5-10 seconds: GET /v2/requests/status/{request_id}
4. When status is "COMPLETED", download from output.media_url

Tip: Use X-Webhook-URL header to get a callback instead of polling.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-turbo-v2-5/v1/text-to-speech' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"text": "Hello from Pixazo. This is a text to speech demo.", "voice_id": "JBFqnCBsd6RMkjVDRZzb"}'

Callback Payload (success)

{
  "request_id": "elevenlabs-turbo-v2-5_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-turbo-v2-5",
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-08-01T09:14:16.102Z",
  "completed_at": "2026-08-01T09:14:22.870Z"
}

Failure callback shape

{
  "request_id": "elevenlabs-turbo-v2-5_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-turbo-v2-5",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on each status change.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Voices

32 languages. Use any voice id from your ElevenLabs voice library.

Request Headers

Response Handling

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED (or ERROR), then download output.media_url.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-turbo-v2-5_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-turbo-v2-5_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-turbo-v2-5",
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-08-01T09:14:16.102Z",
  "completed_at": "2026-08-01T09:14:22.870Z"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

Pricing

Billed at $0.05 per 1,000 characters of input text, rounded up to the next whole 1,000. You are charged for the text you submit, not the length of the audio produced.

Failed requests are not billed.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-voice-changer/v1/speech-to-speech' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"audio_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/doc-assets/audio/speech-17s.mp3", "voice": "Adam"}'

Callback Payload (success)

{
  "request_id": "elevenlabs-voice-changer_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-voice-changer",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Failure callback shape

{
  "request_id": "elevenlabs-voice-changer_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-voice-changer",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on each status change.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

• audio_url must be a public http(s) URL that stays reachable until the request has been processed. The recording is at most 5 minutes long and 32 MB.
• The model is fixed by the URL. Do not send a model_id or a voice_id — both are rejected. Choose the voice by name with voice.

Voices

Choose one of ElevenLabs' preset voices by sending its name, exactly as written in the Voice column, as voice. Custom, cloned and Voice Library voices are not available.

Voice settings

Every field is optional. Send only the ones you want to change for this request.

{
  "audio_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/doc-assets/audio/speech-17s.mp3",
  "voice": "Adam",
  "voice_settings": {
    "stability": 0.4,
    "similarity_boost": 0.8
  }
}

Input recording

• The recording is downloaded when the request is processed, so the URL must stay reachable until then.
• It can be at most 5 minutes long and 32 MB.
• MP3, WAV or M4A is recommended. The length of these formats is read when you submit, so a recording over 5 minutes is rejected straight away and the hold placed on your balance matches the recording. For other formats the hold covers the 5-minute maximum until the converted audio has been measured.

Request Headers

Response Handling

Most parameter problems are caught when you submit and returned as a 400, before any hold is placed on your balance. A few checks run after the request is accepted and are reported through the status endpoint as status: "ERROR", with a message in error saying what to fix. Failed requests are not billed — the hold is released in full.

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED (or FAILED/ERROR), then download the converted audio from output.media_url.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-voice-changer_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-voice-changer_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-voice-changer",
  "error": null,
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-09-15T10:40:08.748Z",
  "completed_at": "2026-09-15 10:40:36"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or FAILED/ERROR (failure).

How billing is measured

Billing is measured on the length of the converted audio, rounded up to the next whole minute — not on how long processing takes. A hold sized from the length of audio_url is placed when the request is submitted; if that length cannot be read, the hold covers the 5-minute maximum. The hold is reduced to the real cost once the converted audio has been measured, and released in full if the request fails.

The current rate is shown in the pricing panel on this page.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Webhook Headers

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-dubbing/v1/dub' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  -H 'X-Webhook-Mode: terminal' \
  --data-raw '{"source_url": "", "target_lang": "de"}'

Callback Payload (success)

{
  "request_id": "elevenlabs-dubbing_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-dubbing",
  "output": {
    "media_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/elevenlabs-dubbing/019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx.mp4",
    "media_type": "video/mp4"
  },
  "duration": 15.86
}

Failure callback shape

{
  "request_id": "elevenlabs-dubbing_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-dubbing",
  "error": "Couldn't extract audio from the downloaded file.",
  "output": null
}

Delivery semantics

• terminal mode (default): one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on every poll.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 OK within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Minimum request

Only source_url and target_lang are required. Everything else has a working default.

{
  "source_url": "https://storage.googleapis.com/eleven-public-cdn/audio/marketing/nicole.mp3",
  "target_lang": "es"
}

Source Media & Languages

Source requirements

• source_url must be a public http(s) URL reachable without authentication — no signed-header or private-bucket URLs.
• The file must contain an audible speech track. Silent video, music-only audio and files whose audio stream cannot be read are rejected.
• A video source returns dubbed video with the original picture intact (video/mp4). An audio source returns audio (audio/mpeg).

Limits

To dub media longer than 600 seconds, split it into segments and submit each segment separately — or use start_time and end_time to dub only the portion you need.

The length check is best-effort: it depends on reading the duration out of the file at source_url. If that duration cannot be read, the job is not blocked — it is dubbed and billed normally. Keep your sources within 600 seconds rather than relying on the check to stop them.

Processing time

Processing time scales with media length. Measured: about 12 seconds for 15 seconds of audio, and about 22 seconds for 16 seconds of video.

Supported languages

This endpoint runs ElevenLabs dubbing_v1, which supports the 86 languages below. Give the code in target_lang (and optionally in source_lang). Use the base language tag — region-qualified variants such as pt-BR, es-MX or en-GB are not supported:

Request Headers

Response Handling

Error Responses

{
  "request_id": "elevenlabs-dubbing_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-dubbing",
  "error": "Couldn't extract audio from the downloaded file.",
  "output": null
}

A failed dub is reported via the status endpoint (or Webhook) with status: "ERROR", an error message and output: null. The messages you are most likely to see:

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED (or FAILED/ERROR).

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-dubbing_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response (video source)

{
  "request_id": "elevenlabs-dubbing_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-dubbing",
  "output": {
    "media_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/elevenlabs-dubbing/019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx.mp4",
    "media_type": "video/mp4"
  },
  "duration": 15.86
}

An audio source returns the same envelope with media_type of audio/mpeg and an audio media_url. output.media_url is a Pixazo-hosted URL. The top-level duration is the length of the source media in seconds — it is what the per-minute charge is computed from.

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or FAILED/ERROR (failure).

Typical Workflow

1. POST https://gateway.pixazo.ai/elevenlabs-dubbing/v1/dub with source_url and target_lang; read request_id from the 202 response.
2. Poll GET https://gateway.pixazo.ai/v2/requests/status/{request_id} — or supply X-Webhook-URL on submit and skip polling.
3. When status is COMPLETED, download output.media_url.
4. Keep no more than 3 dubbing jobs in flight at once.

Pricing

Billed at $0.50 per minute of source media, rounded up to the next whole minute. Billing is measured on the source length, not on the length of the output, and a 15-second clip therefore bills one full minute.

Media longer than 600 seconds is rejected before anything is sent to ElevenLabs — the job ends as ERROR on the first status poll and nothing is charged — so the most expensive possible single request is $5.00.

Authentication

All requests require an API key passed via header.

Webhook (Optional)

Instead of polling, you can receive a Webhook callback when the request reaches a terminal state. Provide a Webhook URL via header on the submit request.

Example: enable Webhook

curl -X POST 'https://gateway.pixazo.ai/elevenlabs-multilingual-v2/v1/text-to-speech' \
  -H 'Content-Type: application/json' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY' \
  -H 'X-Webhook-URL: https://your-server.com/webhook' \
  --data-raw '{"text": "Hello from Pixazo. This is a text to speech demo.", "voice_id": "JBFqnCBsd6RMkjVDRZzb"}'

Callback Payload (success)

{
  "request_id": "elevenlabs-multilingual-v2_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-multilingual-v2",
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-08-01T09:14:16.102Z",
  "completed_at": "2026-08-01T09:14:22.870Z"
}

Failure callback shape

{
  "request_id": "elevenlabs-multilingual-v2_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "elevenlabs-multilingual-v2",
  "error": "Description of the failure"
}

Delivery semantics

• terminal mode: one Webhook callback when the request is COMPLETED or ERROR.
• sync mode: a Webhook callback on each status change.
• Callbacks are idempotent on request_id — de-duplicate on it.
• Respond 200 within a few seconds; the Webhook endpoint must be HTTPS.

Request Parameters

Voices

29 languages. Use an ElevenLabs voice id in voice_id.

Request Headers

Response Handling

Retrieving Results

Poll the status endpoint with the request_id from the submit response until status is COMPLETED (or ERROR), then download output.media_url.

curl 'https://gateway.pixazo.ai/v2/requests/status/elevenlabs-multilingual-v2_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx' \
  -H 'Ocp-Apim-Subscription-Key: YOUR_SUBSCRIPTION_KEY'

Completed response

{
  "request_id": "elevenlabs-multilingual-v2_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "elevenlabs-multilingual-v2",
  "output": {
    "media_url": [
      "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/{request_id}/output.mp3"
    ],
    "media_type": "audio/mpeg"
  },
  "created_at": "2026-08-01T09:14:16.102Z",
  "completed_at": "2026-08-01T09:14:22.870Z"
}

Response Fields

Status Values & Flow

QUEUED → PROCESSING → COMPLETED (success) or ERROR (failure).

How billing is measured

Billing is measured on the input text you submit, in blocks of 1,000 characters rounded up to the next whole block — not on the length of the audio produced. The current rate is shown in the pricing panel on this page.

Failed requests are not billed.