Add Pinyin & Furigana to Subtitles - Reading Aid Guide 2026 | AllSubConverter
Why Add Reading Aids to Subtitles?
Chinese and Japanese subtitles are powerful learning tools, but they share a hurdle: you cannot always tell how a character is pronounced just by looking at it. A Chinese character can have several readings depending on context, and a Japanese kanji can be read in multiple ways. Reading aids solve this. Pinyin tells a learner of Mandarin how to pronounce a Chinese character, while furigana shows the reading of a Japanese kanji in hiragana.
The Subtitle Annotator uses AI to add these reading aids automatically. You upload a subtitle file, choose the annotation mode, and the tool processes each cue to generate the readings, then exports a file you can use for study or karaoke.
Pinyin, Furigana, and Romaji Explained
| Mode | Language | What It Adds |
|---|---|---|
| Pinyin | Chinese | Hanyu Pinyin romanization with tone numbers (e.g. ni3 hao3) |
| Furigana | Japanese | Hiragana readings placed above or beside kanji |
| Romaji | Japanese | Romanized reading of Japanese (e.g. konnichiwa) |
Step-by-Step: Annotating Subtitles
- Upload your subtitle file. The annotator accepts SRT, VTT, ASS, SSA, LRC, SBV, TTML, SCC, STL, and SMI files.
- Enter your OpenAI API key. The key stays in your browser and is sent only to OpenAI, never to our servers.
- Choose the annotation mode. Pick Pinyin for Chinese content, or Furigana/Romaji for Japanese content.
- Pick the reading position. Above Text places the reading on its own line. Inline puts it in parentheses right after the text.
- Click Annotate. The AI processes each cue and adds the readings. A progress bar tracks the work.
- Download. Export as SRT, VTT, or plain text (annotations only).
Use Cases
- Language learning. Add pinyin to Chinese drama subtitles or furigana to anime subtitles for reading practice.
- Karaoke. Add romaji to Japanese song lyrics so non-readers can sing along phonetically.
- Study materials. Build annotated subtitle files for language classes or self-study.
- Reading support. Help viewers who speak a language but cannot read all the characters by providing pronunciation guides.
A common workflow is to first translate or convert the subtitle file, then annotate it for reading, and finally preview the result to make sure the readings line up with the on-screen timing.
Accuracy Notes
The AI uses surrounding context to choose the right reading, which handles most cases well. Chinese has many homophones, though, and Japanese kanji can have multiple readings, so uncommon proper nouns or rare characters may need a manual review. For ambiguous cases, GPT-4o tends to choose more accurate readings than the mini model.
Frequently Asked Questions
Do I need an API key?
Yes. The tool uses your own OpenAI API key. Get one at platform.openai.com. The key stays in your browser and goes only to OpenAI.
Can I annotate Korean subtitles?
The tool is designed for Chinese and Japanese. You can try romaji mode for Korean, but accuracy may vary.
What output formats are available?
SRT, VTT, or plain text. The plain text option contains only the annotations, useful for study notes.