# Convert Chinese bank statements to Excel

Tallyscope reads Chinese column headers, dates and numbers in your browser and checks every row against the running balance before you export to Excel or CSV.

Canonical page: https://tallyscope.app/bank-statement-to-excel-chinese | Last updated: 2026-11-07

What Tallyscope reads in Chinese (中文) statements. The site is in English; no bank names are used or implied.

## Chinese bank statements to Excel

Chinese statements put the year first, write column titles without spaces and may use either simplified or traditional characters. The file is read in your browser tab and every row is checked against the printed running balance. The Chinese-specific parts follow.

## Chinese column headers the reader recognises

| Column | Chinese words |
| --- | --- |
| Date | 日期, 交易日期 |
| Description | 摘要 |
| Debit (money out) | 借方, 借方发生额 |
| Credit (money in) | 贷方, 贷方发生额 |
| Amount | 交易金额 |
| Balance | 余额 |

The statement-level labels **期初余额** and **期末余额** are recognised as the opening and closing balance. Matching ignores case, accents and punctuation, and the lists hold more words than shown. An unrecognised header can be assigned by hand in the column mapper.

## A Chinese sample, row by row

The figures below are synthetic and use Chinese conventions for headers, dates and numbers. The opening label 期初余额 is the statement's starting balance. The same table, saved as a CSV, is read by the real parser in our automated tests, so what follows is what the reader actually does with it.

| 日期 | 摘要 | 借方 | 贷方 | 余额 |
| --- | --- | --- | --- | --- |
| 期初余额 | 1,500.40 |  |  |  |
| 2026-01-02 | 工资 |  | 2,944.00 | 4,444.40 |
| 2026-01-15 | 房租 | 895.50 |  | 3,548.90 |
| 2026-01-20 | 超市 | 112.74 |  | 3,436.16 |
| 2026-01-28 | 退款 |  | 24.04 | 3,460.20 |

- 2026-01-02 is read as 2 January 2026. 2,944.00 in the 贷方 column becomes **+2,944.00**, and the printed balance 4,444.40 is 4,444.40, which equals 1,500.40 plus 2,944.00.
- 2026-01-15 is read as 15 January 2026. 895.50 in the 借方 column becomes **-895.50**, and the printed balance 3,548.90 is 3,548.90.
- 2026-01-20 is read as 20 January 2026. 112.74 in the 借方 column becomes **-112.74**, and the printed balance 3,436.16 is 3,436.16.
- 2026-01-28 is read as 28 January 2026. 24.04 in the 贷方 column becomes **+24.04**, and the printed balance 3,460.20 is 3,460.20.

Opening 1,500.40, plus the four movements (net +1,959.80), gives 3,460.20, the last printed balance 3,460.20. Every row agrees with the running balance, so this statement is shown as Verified.

## Money out and money in in Chinese statements

This sample uses two money columns, a layout found in many Chinese-language statements: 借方 holds money out and 贷方 holds money in, and each amount is printed without a sign because the column carries the direction. The reader takes the direction from the column and then checks it against the balance; if a column were swapped, the balance would not follow and the statement would not be marked Verified.

## Chinese OCR at a glance

| Model code | `chi_sim` |
| --- | --- |
| Writing system | Han characters |
| Bundled English model | Cannot read Chinese letters; a scanned Chinese statement needs the chi_sim model |
| Text PDF, spreadsheet, bank file | No OCR and no model download |

## Dates in Chinese statements

The form 2026年1月15日 is read as 15 January 2026, as is the ISO form 2026-01-15. The Republic of China calendar (民國 or 民国 followed by a year) is also converted, and a bare year in the 100-199 range, which cannot be a Gregorian year, is converted and flagged for confirmation.

Examples that are read correctly: `2026年1月15日`, `2026-01-15`.

## Numbers and signs

Amounts such as 1,234.56 use a comma for thousands and a point for decimals and are read directly; full-width digits and punctuation are normalised first.

Examples that are read correctly: `1,234.56`. The running balance is the final judge of every sign.

## Scanned Chinese statements need the Chinese OCR model

The app lists a Simplified Chinese model (chi_sim). There is no Traditional Chinese model: for locales such as zh-TW or zh-HK the app suggests none, so a scan in traditional characters may read poorly. With the OCR language set to English only, nothing is downloaded; in automatic mode (the default) a poor read can trigger up to three trial model downloads from the jsDelivr CDN (cdn.jsdelivr.net), plus any language remembered from this session or detected earlier, and no statement content is ever sent. The header vocabulary covers traditional characters too.

The file stays in your tab ([every network use is listed here](https://tallyscope.app/convert-bank-statement-without-uploading)).

## What is specific to Chinese

Header matching for unspaced scripts looks for known words inside longer cells, so a title such as 借方发生额(元) (debit turnover, in yuan) still resolves to the debit role. Simplified and traditional forms are both in the list, for example 贷方 and 貸方.

余额 is the running balance, 摘要 the description, and 期初余额 and 期末余额 are recognised as opening and closing balances, so the check can run from the printed figures.

## Convert a Chinese statement

Add the file to [Tallyscope](https://tallyscope.app/) (10 files at once are free), read each statement's status, fix flagged rows on the Review tab, then export: CSV is free for the first 25 rows; unlimited CSV and Excel, PDF, JSON, OFX, QuickBooks CSV, Xero CSV and a reconciliation report (PDF, plus an Excel sheet) come with the Day pass or Pro ([pricing](https://tallyscope.app/pricing)). Several Chinese and non-Chinese files can be [merged](https://tallyscope.app/merge-bank-statements) into one table.

No accuracy percentage is claimed: layouts vary, and the arithmetic check says whether a given statement reconciled ([how verification works](https://tallyscope.app/how-verification-works)).

## Frequently asked questions

### Does Tallyscope read Chinese bank statements?

It recognises Chinese column headers, balance labels, dates and number formats, and checks the rows against the running balance. The website itself is only in English.

### Do I need to download an OCR model for Chinese?

For scanned images, yes: Chinese script cannot be read by the bundled English model, so a Chinese model (chi_sim) is downloaded from the jsDelivr CDN (cdn.jsdelivr.net) when you choose that language or auto-detect needs it. Text PDFs and spreadsheets need no OCR.

### Are my Chinese statements uploaded?

No. Files are read in your browser tab. The only OCR download is a language model file from the jsDelivr CDN, requested only when a scan needs a non-English model; no statement content is sent with it.

## Related guides

- [Bank statement OCR](https://tallyscope.app/bank-statement-ocr): In-browser OCR, repairs only when arithmetic proves them
- [Scanned statement to Excel](https://tallyscope.app/scanned-bank-statement-to-excel): Photos and scanned PDFs read by in-browser OCR
- [Finnish bank statements to Excel](https://tallyscope.app/bank-statement-to-excel-finnish): Finnish headers, dates and numbers
- [Estonian bank statements to Excel](https://tallyscope.app/bank-statement-to-excel-estonian): Estonian headers, dates and numbers
- [Multi-currency statements](https://tallyscope.app/multi-currency-bank-statements): Many currencies, one reporting currency, live rates