# Chinese speech-to-text API (中文)

> KolWrite transcribes Mandarin Chinese audio and video with one API call: send "language": "zh" and get Chinese characters in segments with timestamps and speaker labels, as JSON, TXT or SRT. Oracle models translate Chinese into English or 99 other languages.

Last updated: 2026-10-08 · HTML: https://console.kolwrite.com/speech-to-text/chinese

## Transcribe Chinese with one request

Set `language` to `zh` and start a job. Poll `GET /transcribe/{jobId}` for the result, or pass a `webhook_url`.

```bash
curl https://app.kolwrite.com/transcribe \
  -H "x-api-key: $KOLWRITE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"audio_url": "https://example.com/meeting-zh.m4a", "model": "oracle", "language": "zh", "include_diarization": true, "output_format": "json"}'
```

Translate Chinese audio into English in the same job (Oracle models only):

```bash
curl https://app.kolwrite.com/transcribe \
  -H "x-api-key: $KOLWRITE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"audio_url": "https://example.com/meeting-zh.m4a", "model": "oracle_mini", "language": "zh", "translate_to": "en", "output_format": "srt"}'
```

## Chinese at a glance

| Property | Details |
| --- | --- |
| API code | `zh` |
| Native name | 中文 |
| Writing direction | Left to right |
| Automatic detection | Yes: omit `language` or send `"auto"` |
| Translate Chinese audio into | Any of the 100 supported languages, on Oracle models |
| Translate other languages into Chinese | Yes: `translate_to: "zh"` on Oracle models |
| Output | JSON (segments, word timings, speakers), TXT, SRT |
| Price | US$0.48 (Base) to US$3.60 (Oracle Max) per audio hour; US$1.44 on Oracle |

## What to know about Chinese transcription

- The API has one Chinese code, `zh`, and no script parameter. If your product needs Simplified or Traditional characters consistently, convert the transcript afterwards (for example with OpenCC).
- Chinese is written without spaces, so a "word" in the timestamps may be a single character or a short token. Build subtitles and highlights by time ranges, not by word counts.
- Cantonese and other Chinese languages are not listed separately. Test Cantonese, Hokkien or Wu recordings before relying on them.

## Where teams use Chinese transcription

- Meeting and interview transcripts for teams working with China.
- Chinese subtitles for video and e-learning (`srt`).
- Translating Mandarin content to English for research or due diligence.

## Choosing a model

Start with `oracle` (US$1.44 per hour) for production Chinese audio. Use `base` (US$0.48) for clean, single-speaker recordings at volume, `oracle_mini` (US$0.96) for the cheapest translation, and `oracle_pro` or `oracle_max` for noisy or difficult audio. KolWrite does not publish a Chinese accuracy figure for this API, so test a sample of your own recordings before committing. Full comparison: [models and prices](https://console.kolwrite.com/models).

## Questions and answers

### Does KolWrite return Simplified or Traditional Chinese?

There is no parameter for it. Convert to the script you need in post-processing.

### Is Cantonese supported?

The supported list has one Chinese code, `zh`, meant for Mandarin. Cantonese is not listed separately; test it on your own audio.

### Can I translate Chinese audio to English?

Yes, with an Oracle model and `translate_to: "en"`.

## Related pages

- [Japanese speech-to-text](https://console.kolwrite.com/speech-to-text/japanese)
- [Korean speech-to-text](https://console.kolwrite.com/speech-to-text/korean)
- [English speech-to-text](https://console.kolwrite.com/speech-to-text/english)
- [Developers & product teams](https://console.kolwrite.com/use-cases/developers)
- [Language research & linguistics](https://console.kolwrite.com/use-cases/researchers)
- [Supported languages](https://console.kolwrite.com/languages/)

---
Get an API key: https://console.kolwrite.com/auth · API documentation: https://console.kolwrite.com/public_api_docs · llms.txt: https://console.kolwrite.com/llms.txt