Chinese Short Drama into English: One Cast, Nine Languages
Take a Chinese short-drama script into English, Spanish, Japanese and more with the same character voices, with real length and cost data for seven versions.
VoiceSmiths TeamListen to this article
AI narration · about 5 min
Short answer: do not translate first and cast later. Cast the Chinese script once, then adapt that cast into each target language, so every character keeps one voice everywhere. In the VoiceSmiths Audio Drama Studio: paste the Chinese script, Find the characters, check the cast, pick a target language and click Adapt (same cast) with "keep the same voices" on. Each actor then performs the new language. The studio adapts into English, Spanish, Portuguese, Japanese, Korean, Indonesian, French and German.
We did exactly this with one web-novel scene and kept all seven versions. The numbers below come from those files.
"Same cast" means two different things
1. The same voice speaking another language. The heroine sounds like herself in Chinese and in Spanish. This is what audiences who watched the Chinese original expect, and what keeps a character recognisable across a multilingual release.
2. A native voice per language, same character. You bind each character to a voice that is native to the target market. Better when the audience never saw the original.
The studio defaults to the first (the checkbox is "Keep the same voices (each actor speaks the other language)"); untick it to recast per language.
If what you have is a finished live-action video and you want the original actors' voices speaking English, that is video dubbing, a different tool. See dubbing short dramas for overseas markets and video dubbing.
The workflow
- Paste the Chinese script (up to 8,000 characters) and click Find the characters. Narration and dialogue are separated and each character gets a voice. How attribution works, and how to help it, is in turning a web novel chapter into a full-cast audio drama.
- Fix the cast in Chinese first. Every correction you make here carries into every language.
- Pick a target language and click Adapt (same cast). The adaptation rewrites rather than translates word for word: idioms, forms of address and genre tropes are carried over in a form the target audience reads naturally.
- Read the adapted script. Character names can be edited in the cast list (we kept pinyin, Shen Qingci, in the Latin-script versions). You can change any line's speaker and emotion tag; the wording itself comes from the adaptation, so have a native speaker listen to the preview before rendering a whole series.
- Preview the opening, then render each language separately. Each render comes with its own MP3, SRT subtitles and script.
One scene, seven languages: the real numbers
The scene is Reborn at My Engagement Party, a five-voice web-novel opening. Same five voices in every version:
| Version | Characters | Audio | Credits to render |
|---|---|---|---|
| Chinese (original) | 811 | 3:10 | 57 |
| English | 2,979 | 3:36 | 209 |
| Spanish | 3,016 | 3:44 | 212 |
| Portuguese | 2,910 | 3:45 | 204 |
| Indonesian | 3,046 | 3:57 | 214 |
| Korean | 1,552 | 3:53 | 109 |
| Japanese | 1,396 | 4:01 | 98 |
Credits are at the studio's 70 credits per 1,000 characters.
Three things in that table matter when you plan a release:
- Every version runs longer than the Chinese one, by 14% (English) to 27% (Japanese). If the audio has to sit under existing video cuts, plan for the extra time, or trim the adapted lines.
- Cost follows characters, not minutes. The Latin-script versions need about 3.7 times as many characters as the Chinese for the same scene, so they cost about 3.7 times as much to render. Japanese and Korean sit in between.
- The cast does not drift. Keep the same voice per character from episode to episode and episode two sounds like episode one in every language, which is the hard part with a separate human dubbing team per market.
Listen
The English version:
The same cast in Spanish:
And in Japanese:
The Chinese original, Portuguese, Korean and Indonesian versions are in the theater, each with its script to read along.
Limits worth knowing
- It produces audio, not video. No lip-sync; the SRT is there for your editor.
- The adaptation is AI-written. It handles names, address and idiom well enough to publish after review, not without one.
- No music or sound effects. Add them in post.
FAQ
Do I need to translate the script first?
No. Paste the Chinese script; the adaptation produces each target-language script from it.
Which languages are supported?
The source can be Chinese or English. Adaptations: English, Chinese, Spanish, Portuguese, Japanese, Korean, Indonesian, French and German.
Can I keep a real actor's voice from my video?
Not in the audio drama studio, whose voices are AI-designed characters. To keep the original actors' voices from a finished video, use video dubbing.
How much does a 90-second episode cost?
About 800 Chinese characters of script is 56 credits in Chinese; the English version of the same episode is roughly 3,000 characters, about 210 credits. See pricing for what credits cost on each plan.
Can I use the result commercially?
Audio you generate is yours and may be used commercially under the terms of service.