ElevenLabs Models Updated: New Speed Control Parameter for More Flexible Voice Adjustment

2026-02-10
UnifiedTTS Team

We are excited to announce that ElevenLabs speech synthesis models now support the speed control parameter! Following the stability, similarity boost, and style parameters, this new feature gives you even more precise control over your speech synthesis output.

New Parameter: Speed Control

Parameter Name: speed

Valid Range: 0.7 - 1.2

Default Value: 1.0 (normal speed)

Speed Range Guide

Value Effect Use Case
0.7 Slowest speed Tutorials, listening practice, content requiring careful attention
0.8-0.9 Slower speed Audiobooks, news broadcasting, formal occasions
1.0 Normal speed Daily conversation, general use (default)
1.1 Faster speed Quick updates, energetic content
1.2 Fastest speed Efficient delivery, fast-paced scenarios

How to Use

Control speech rate by adding the speed parameter when calling the TTS API:

{
  "model": "eleven_turbo_v2_5",
  "voice": "21m00Tcm4TlvDq8ikWAM",
  "text": "This is a test message demonstrating the speed control feature.",
  "speed": 0.8
}

Examples in Different Languages

English at normal speed (speed: 1.0):
The quick brown fox jumps over the lazy dog.

English at slower speed (speed: 0.8):
The quick brown fox jumps over the lazy dog.

Chinese at normal speed (speed: 1.0):
人工智能正在改变我们的生活方式。

Chinese at slower speed (speed: 0.8):
人工智能正在改变我们的生活方式。

Supported Models

The speed control parameter is available for the following ElevenLabs models:

  • eleven_multilingual_v2 - Multilingual model
  • eleven_turbo_v2_5 - Turbo model
  • eleven_flash_v2_5 - Flash model

Combining with Other Parameters

ElevenLabs offers a complete set of voice control parameters that you can combine:

Parameter Range Purpose
speed 0.7-1.2 Control speech rate
stability 0.0-1.0 Control voice stability
similarity_boost 0.0-1.0 Enhance similarity to original voice
style 0.0-1.0 Control style exaggeration

Combined Usage Example

{
  "model": "eleven_multilingual_v2",
  "voice": "21m00Tcm4TlvDq8ikWAM",
  "text": "Welcome to our speech synthesis service.",
  "speed": 0.9,
  "stability": 0.7,
  "similarity_boost": 0.5,
  "style": 0.3
}

Typical Use Cases

Education & Training

Use slower speeds (0.7-0.8) to give learners enough time to understand and take notes.

Audiobooks

Adjust speed based on content type: normal speed for novels, slightly slower for technical documents.

News & Updates

Use faster speeds (1.1-1.2) to deliver more information in limited time.

Game Voiceovers

Adjust speed based on character personality: faster for energetic characters, slower for calm ones.

Customer Service

Adjust based on scenario: normal speed for urgent issues, slightly slower for detailed explanations.

Important Notes

  1. Range Limitation: ElevenLabs speed parameter range is 0.7-1.2. Values outside this range will be automatically adjusted to valid limits.

  2. Unified API: If you're using the unified TTS interface with a 0.5-2.0 range, the system will automatically map values to ElevenLabs' valid range.

  3. Quality Impact: Extreme speeds may slightly affect audio quality. For best results, we recommend staying within the 0.8-1.1 range.


Try the ElevenLabs speed control feature today and make your voice content more expressive!