Skip to main content

Text-to-Speech (Audio Generation) with ElevenLabs

What It Does​

The Text-to-Speech feature automatically converts text messages into natural-sounding voice notes that are sent as WhatsApp audio messages inside your CRM conversations using ElevenLabs API.
This allows you to send voice messages from automation workflows — no need to record audio manually.

You can choose from a wide variety of AI voices and even clone custom voices using ElevenLabs.
It's perfect for creating personalized audio responses at scale.

How to Activate Text-to-Speech​

Steps:​

  1. Go to you locations and look for WhatsApp menu item in the sidebar.

  2. Once you click on WhatsApp menu item, Go to Settings.

  3. Scroll down to the Text-to-Speech (Audio Generation) section.

  4. Enter your ElevenLabs API Key and Toggle Enable Text-to-Speech and click Save.

How to Use Text-to-Speech​

To send a voice note, use this format in any WhatsApp message field (manual messages, workflows, or automations):

Message Format:​

[AUDIO] [voice_id] Your text message here

Example:​

[AUDIO] [XrExE9yKIg1WjnnlVkGX] In the ancient land of Eldoria, where skies shimmered and forests whispered secrets to the wind, lived a dragon named Zephyros.

Format Breakdown:​

  • [AUDIO] - Trigger keyword (required)
  • [voice_id] - Your ElevenLabs Voice ID (required)
  • Text after the voice ID - The message that will be converted to speech

Where You Can Use It:​

  • Manual messages in conversations
  • Automated workflow responses
  • Trigger-based messages
  • Scheduled broadcasts
  • Custom values and variables

Using Variables:​

You can also combine it with custom fields:

[AUDIO] [XrExE9yKIg1WjnnlVkGX] Hi \{\{contact.first_name\}\}, your appointment is confirmed for \{\{contact.appointment_date\}\}.

How to Get Your ElevenLabs API Key​

  1. Go to ElevenLabs.
  2. Log in or create an account.
  3. Navigate to your Profile Settings.
  4. Click Create API Key.
  5. Important: When creating the API key, make sure to enable these permissions:
    • ✅ Text-to-Speech
    • ✅ Voices (to access voice library)
  6. Copy the generated key.
  7. Paste it into the Text-to-Speech settings.
API Key Permissions

Make sure to enable Text-to-Speech and Voices permissions when creating your ElevenLabs API key, otherwise the feature won't work.

How to Find Voice IDs​

Every voice in ElevenLabs has a unique Voice ID that you'll need to use in your messages.

Steps to Get a Voice ID:​

  1. Go to your ElevenLabs Dashboard.
  2. Browse the Voice Library or use your custom cloned voices.
  3. Click on any voice you want to use.
  4. Copy the Voice ID (it looks like: XrExE9yKIg1WjnnlVkGX).
  5. Use this Voice ID in your message format (see below).
Voice Library

ElevenLabs offers hundreds of pre-made voices in multiple languages. You can also clone your own voice for brand consistency.

FAQs​

Does this consume a lot of ElevenLabs credits?

Usage depends on the length of your text. ElevenLabs charges based on characters converted. The free plan includes a limited number of characters per month, while paid plans offer higher limits.

Can I use this in automation workflows?

Yes! You can use the [AUDIO] [voice_id] format in any message field within workflows, triggers, or automations.

Can I use my own cloned voice?

Yes. If you have created a custom voice clone in ElevenLabs, simply copy that voice's ID and use it in the message format.

Do I need a paid ElevenLabs account?

No, the feature works with both free and paid ElevenLabs accounts. However, paid plans offer more characters and faster processing.

What happens if I use the wrong Voice ID?

If the Voice ID is invalid or not accessible with your API key, the message will fail to send or send as text instead.

Can I use different voices for different messages?

Yes! Simply use different Voice IDs for each message. You can switch voices as needed.

What languages are supported?

ElevenLabs supports multiple languages. Check the ElevenLabs Voice Library for available languages and voices.

Is there a limit to message length?

WhatsApp voice notes have a practical limit. We recommend keeping messages under 2-3 minutes (approximately 300-450 words) for best results.
Let's connect