Lesson examples
Single sentences for learners to repeat, the use our note names. One example per line gives one caption per example.

Tongtong supplies most of Soft Mandarin, a recipe we kept: 80 percent of the mix behind Ruoxi, and our note calls the result gentler than either voice alone. Her own use is short lesson examples, single sentences a learner can repeat.
Hear it
About this voice
Soft Mandarin is unusual among our recipes because both voices are Mandarin. Ruoxi in front sets the pronunciation; Tongtong at 80 percent supplies most of the timbre. Our listening note is short: gentler than either alone.
Lesson examples are short by nature, a sentence read once so a learner can repeat it. Captions cut at real pauses fit that pattern: one example, one cue. Slow the speed for practice and the cues stretch with the audio.
Length the engine likes. The published guidance puts the sweet spot at roughly 100 to 200 tokens, a few sentences per line.
Where we would use it
Single sentences for learners to repeat, the use our note names. One example per line gives one caption per example.
Word lists and short phrases, read one at a time so a learner can say each one back. Slow the speed for practice.
The kept recipe with Ruoxi in front, gentler than either alone by our note. Try it for longer lesson passages.
Blends
Soft Mandarin is the recipe we kept, with Tongtong as the second voice. The cards below put her in front instead, and each one is ready to hear in the widget.
Reverse of Soft Mandarin: Tongtong in front, Ruoxi behind. The kept version runs the other way, so this card is a fresh pairing to hear.
50% Ruoxi
A pairing to try for lessons. Nora is our lesson-introduction voice; listen for how 60 percent of her changes an example sentence.
60% Nora
A pairing to try. Imogen is our British female voice with a steady pace and a courseware note; listen for a courseware timbre on Mandarin.
60% Imogen
A pairing to try. Delaney supplies the throaty edge in five of our kept Mandarin blends; listen for how it sounds on a lesson.
40% Delaney
Two Mandarin voices blended, gentler than either alone.
Ruoxi 20% + Tongtong 80%
A fixed sample at this mix, rendered once and cached. No credits used.
Every blend works the same way: the first voice sets the pronunciation and pacing, the second voice lends only its timbre, and you choose the ratio between 10 and 90 percent. See the three steps on the voice library page
At a glance
| Language | Chinese |
|---|---|
| Voice | Female |
| Captions | Synced captions |
| Cost | Up to 10 minutes of audio per credit |
| Formats | MP3, plus WAV on paid plans |
| Commercial use | Included on every plan |
| Preview | Free, no sign-up |
FAQ
She is 80 percent of it, with Ruoxi in front for pronunciation. Our note calls the blend gentler than either alone.
Yes. Put each example on its own line and end it with a period; the caption cue is anchored to the pause the engine makes there, with lines capped at 16 characters.
Latin text is split off and read as English, which suits English terms in a lesson. For Mandarin words, write the characters and the voice pronounces them in Mandarin.
Play the three samples above, then try Soft Mandarin in the widget. Both preview free and need no account.
Similar voices
Paste a script, keep Tongtong or blend a second voice, and download the MP3 with subtitles.