ElevenLabs#

JSON2Video API integrates with ElevenLabs to provide high-quality AI-powered voiceovers. This integration enables you to generate realistic and expressive speech from text directly within your video creation workflow.

Key Features#

• High-Quality Voice Generation: Leverage ElevenLabs' advanced AI models for natural-sounding voiceovers.
• Voice Variety: Choose from a wide range of pre-existing voices or use custom ElevenLabs voices.
• Simplified Integration: Seamlessly integrate voiceover generation into your video creation process with the Voice element.

How to Use ElevenLabs in JSON2Video#

1. Select the your preferred model: In your voice element, specify elevenlabs or elevenlabs-flash-v2-5 as the model value.
2. Choose a Voice: Provide the name of a voice you want to use in the voice property. For example, "Adam" or "Rachel". See ElevenLabs Voices Library for available voices.
3. Provide Text: Set the text you want the voice to speak in the text property.
4. Configure a Connection (Optional): If you have your own ElevenLabs API key, create a Connection and set the connection ID to use your account.

Credit Consumption#

Using the ElevenLabs models will consume credits from your JSON2Video account. Refer to the Credit Consumption section for more details.

Bring your own ElevenLabs API key#

If you have your own ElevenLabs API key, you can use it to generate voiceovers. Check this page for more information on how to create a Connection to your ElevenLabs account.

Examples#

Basic Example:

{
  "scenes": [
    {
      "elements": [
        {
          "type": "voice",
          "model": "elevenlabs",
          "voice": "Adam",
          "text": "Hello, I am an ElevenLabs generated voiceover."
        }
      ]
    }
  ]
}

Using Your Own ElevenLabs API Key:

{
  "scenes": [
    {
      "elements": [
        {
          "type": "voice",
          "model": "elevenlabs",
          "connection": "my-elevenlabs-connection",
          "voice": "p16ZaTyG1Ks9FQ9LpSun",
          "text": "Hello, world!"
        }
      ]
    }
  ]
}

Example using the ElevenLabs Flash v2.5 model:

ElevenLabs Flash v2.5 model is a faster way to generate voiceovers that also supports a wide range of voices and languages. The credit consumption is the same of the ElevenLabs model. To generate a voiceover using Flash v2.5, use elevenlabs-flash-v2-5 as the value for the property model:

{
  "scenes": [
    {
      "elements": [
        {
          "type": "voice",
          "model": "elevenlabs-flash-v2-5",
          "voice": "Adam",
          "text": "Hello, I am an ElevenLabs Flash v2.5 generated voiceover."
        }
      ]
    }
  ]
}

Customizing ElevenLabs settings#

JSON2Video's Voice element supports the model-settings property to pass custom configuration to ElevenLabs API. You can read the available settings in ElevenLabs's API reference here: https://elevenlabs.io/docs/api-reference/text-to-speech/convert#request

Example using the model-settings property:

{
  "scenes": [
    {
      "elements": [
        {
          "type": "voice",
          "model": "elevenlabs",
          "voice": "p16ZaTyG1Ks9FQ9LpSun",
          "text": "Hello, world!",
          "model-settings": {
            "language_code": "en"
          }
        }
      ]
    }
  ]
}

Changing the ElevenLabs speech speed#

One of the voice_settings is speed that controls the speed of the voiceover. The value is a number between 0.7 and 1.2.

Example changing the ElevenLabs speech speed:

{
  "scenes": [
    {
      "elements": [
        {
          "type": "voice",
          "model": "elevenlabs",
          "voice": "p16ZaTyG1Ks9FQ9LpSun",
          "text": "Hello, world!",
          "model-settings": {
            "voice_settings": {
              "speed": 1.2
            }
          }
        }
      ]
    }
  ]
}