Model Updates: Streamlined Model Selection for Better Voice Synthesis

Today we've made important updates to our supported TTS models, removing legacy versions and adding superior new models. This update aims to simplify your model selection, avoid choice overload, and ensure you're using the best value-for-money solutions.

Update Overview

Removed Models

We've removed the following legacy models:

MiniMax Models

• speech-2.5 series

  • Removed models: speech-2.5-hd-preview, speech-2.5-turbo-preview
  • Reason: Replaced by more powerful speech-2.6-hd and speech-2.6-turbo at the same price point

• speech-01 series

  • Removed models: speech-01-hd, speech-01-turbo
  • Reason: Better alternatives available with speech-02-hd and speech-02-turbo at the same price

ElevenLabs Models

• ELEVEN_MULTILINGUAL_V1

  • Removed model: eleven_multilingual_v1
  • Reason: Replaced by upgraded multilingual model eleven_multilingual_v2 - same price, better performance

• ELEVEN_TURBO_V2

  • Removed model: eleven_turbo_v2
  • Reason: Upgraded to v2.5 version eleven_turbo_v2_5

• ELEVEN_FLASH_V2

  • Removed model: eleven_flash_v2
  • Reason: Upgraded to v2.5 version eleven_flash_v2_5

Added Models

We've added the following high-performance models:

MiniMax New Models

• speech-2.6
  • Performance improvement: More natural voice quality and faster response compared to v2.5
  • Price advantage: Same price as v2.5 with better value
  • Language support: Chinese, English, and other languages

ElevenLabs Upgraded Models

• ELEVEN_TURBO_V2_5

  • Performance improvement: 30% faster response than v2
  • Quality optimization: More natural voice synthesis
  • Pricing unchanged: Same price as v2

• ELEVEN_FLASH_V2_5

  • Ultra-fast: Optimized for low-latency scenarios
  • Cost optimization: Lower cost while maintaining high quality
  • Wide applications: Ideal for real-time conversations, customer service, etc.

Why This Update?

1. Simplified Choices, Less Confusion

As TTS technology evolves rapidly, newer models outperform older versions in every aspect. Keeping too many versions causes confusion: "Should I use v2 or v2.5?"

Our principle: Keep only the best model at each price point to make your choice simpler.

2. Better Value for Money

New models offer at the same or lower price:

• Faster response times
• More natural voice quality
• Better multilingual support
• More stable service quality

3. Unified User Experience

By removing legacy models, all users will use the latest and best solutions, helping us to:

• Provide more consistent service quality
• Focus on optimizing core models
• Better support user needs

Future Plans

We will continue to monitor TTS technology developments and regularly evaluate and update our supported model list to ensure you always have access to:

• 🎯 Best performing models
• 💰 Best value-for-money solutions
• 🚀 Latest technological innovations
• 📊 Clear and simple choices

Thank you for supporting UnifiedTTS!

If you have any questions or suggestions, feel free to contact us via email.