Duan Yu Liu Mai | Duan Yu Six Meridian | Noiz AI

Duan Yu Six Meridian is a Noiz AI voice for gaming and short videos, with a calm and credible tone. Powered by Noiz AI, with realistic voice cloning and emotional TTS, trusted by 2M+ creators.

Try this voice here

😐 让我们一起创造 NOIZ,为世界添上更多色彩。

Emotional Expression

Similar Voices

Explore similar voices by tone, style, or emotion in Noiz’s vast library and try it free today.

Explore More

Samples for Every Scene

Hear how this voice performs across different styles, emotions, and use cases.

Game Character Game characters
Match Commentary Match commentary
YouTube Intro YouTube intros

How It Works

Enter Your Script

Paste your script, then direct the performance with emotion tags like [😊#Happy:7] or presets like warm, tense, playful, and direct. Adjust speed, emphasis, and feel.

Select a Voice

Choose from millions of voices across languages, tones, and use cases, then preview with your own script in real time — not generic samples.

Generate & Export

Generate in one click, preview the result, and refine until it sounds right. Export high-quality MP3 ready for your project — start with a free trial.

Frequently Asked Questions

Is this Duan Yu Liu Mai's real voice?

This is an AI-generated voice inspired by Duan Yu Liu Mai. It is not Duan Yu Liu Mai's real voice, is not authorized or endorsed, and is not affiliated with the original creators or rights holders.

What types of content suit the Duan Yu Six Meridian voice best?

This voice works well for indie games, fan dubs, and narrative shorts where a calm and credible tone is required.

Can I use this voice for commercial projects?

Yes, you can use this AI voice for commercial purposes according to the platform's licensing terms for generated assets.

How natural does the Duan Yu Six Meridian voice sound?

The voice features advanced prosody modeling to reduce robotic artifacts, resulting in a more human-like listening experience.

Does this voice support multiple languages?

Currently, the Duan Yu Six Meridian persona is optimized for English delivery with specific stylistic traits.

What makes this persona different from generic TTS voices?

It offers a distinct character profile with consistent emotional coloring, unlike standard neutral text-to-speech models.

Is the audio output high fidelity?

Yes, the system generates studio-quality audio files suitable for professional broadcasting and streaming platforms.

How do I adjust the speaking speed?

You can control the pacing through the API parameters or the dashboard interface to match your scene requirements.

Ready to Create with This Voice?