MiniMax Speech 2.8 Released: Emotional Tags for More Expressive Speech

2026-02-14
UnifiedTTS Team

We are excited to announce that MiniMax Speech 2.8 series models are now live on UnifiedTTS! This is another major upgrade to MiniMax's speech synthesis technology, bringing the all-new emotional tone tags feature that makes AI-generated speech more vivid, natural, and expressive.

Speech 2.8 Series Models

🎭 speech-2.8-hd: High Definition

Model ID: speech-2.8-hd

Pricing: 600 credits per 10K characters

Features:

  • Highest audio quality output, ideal for professional scenarios
  • Full support for emotional tone tags
  • More nuanced emotional expression

⚡ speech-2.8-turbo: Fast Version

Model ID: speech-2.8-turbo

Pricing: 300 credits per 10K characters

Features:

  • Fast response, ideal for high-concurrency scenarios
  • Full support for emotional tone tags
  • Better value for money

🎉 Major New Feature: Emotional Tone Tags

The biggest highlight of Speech 2.8 is the support for emotional tone tags. You can insert specific tags in your text to control the emotional expression of the speech:

Supported Tone Tags

Tag Effect Example
(laughs) Laughter That's so funny(laughs)
(chuckle) Light laugh That's a good idea(chuckle)
(sighs) Sighing Oh well(sighs), what can I do
(gasps) Surprised gasp Wow(gasps)! That's amazing
(clears throat) Clearing throat Ahem(clears throat), let's begin
(coughs) Coughing Excuse me(coughs), let me continue
(sneezes) Sneezing Achoo(sneezes)! Sorry about that

Usage Examples

The weather is so nice today(laughs), let's go out and play!
This problem is indeed quite complex(sighs), but I'm confident we can solve it.
Wow(gasps)! This news is so shocking(chuckle)!

📊 Comparison with Previous Version

Feature Speech 2.6 Speech 2.8
Audio Quality Excellent Excellent
Response Speed Fast Fast
Emotional Tone Tags ❌ Not Supported ✅ Supported
Price 600/300 credits 600/300 credits

Same price, more powerful features!

💡 Typical Use Cases

🎬 Film & Video Dubbing

Simulate real actors' emotional expressions through tone tags for more natural narration and dubbing.

📚 Audiobooks

Add emotional color to novels and stories, making characters come alive.

🎮 Game NPCs

Make in-game character dialogue more vivid and enhance immersion.

🎙️ Podcast Content

Add emotional variation to host voices for a more authentic sound.

📱 Social Media

Create engaging short video voiceovers to attract more viewers.

How to Use

Simply select the new model when calling the TTS API:

{
  "model": "speech-2.8-hd",
  "voice": "male-qn-qingse",
  "text": "This feature is amazing(laughs)!"
}

Try MiniMax Speech 2.8 today and make your voice content more vivid and engaging!