在日常学习、教学或资料整理中，我们经常需要给中文文字加上拼音标注。
传统方法要么用 Word 插件、要么手工标注，效率低又容易出错。
最近发现一款 AI 拼音标注工具 ，它能自动识别整段中文内容，并精准地为每个汉字加上拼音，只需一键即可生成标注版文本或文档。
支持网页端和桌面端使用，完全免安装，使用体验非常丝滑。
主要功能
- 自动拼音标注：粘贴或上传文本，AI 自动识别并生成拼音。
- 格式智能排版：拼音在汉字上方或括号中显示，排版清晰美观。
- 支持多种输出格式：支持复制、导出为 Word / PDF / TXT 文件。
- 批量处理文本：可一次性对整篇文章、课文或文档进行标注。
- 多音字智能识别：基于 AI 模型自动判断正确读音，准确率高。
- 中英混排识别：自动跳过英文部分，输出干净整齐。
使用步骤
- 双击
pinyin_tools.exe工具即可。 - 将需要标注拼音的文本粘贴到输入框中；或者需要标注拼音的文档导入进来即可。
使用场景
- 教育教学：语文老师准备拼音版课件或试题时的高效助手。
- 儿童学习：家长打印带拼音的故事书、诗词读物。
- 办公文档：需要制作双语或拼音资料的培训讲师、HR 等。
- 语言学习：为国际学生制作带拼音的中文教材。
使用总结
这款 AI 拼音标注工具真正实现了输入中文，秒出拼音，
准确率高、排版美观，对老师、学生和创作者都非常实用。
尤其在中文教育、拼音教学场景下，它能显著提高文档制作效率。
如果你经常需要加拼音，不妨试试看，让 AI 帮你省下大量时间！
现成工具自取，已打包好exe程序：
夸克下载链接：https://pan.quark.cn/s/3d52da2faa62
迅雷下载链接：https://pan.xunlei.com/s/VOcA6WSqXxYeuKLUjqXO3OcvA1?pwd=9txy#
源码
python
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Pinyin Annotation Tool (Windows, Python 3.7.x)
Features:
- wxPython UI: pick .docx/.txt/.pdf files; paste text; live logs
- Add Pinyin above Simplified/Traditional Chinese via Word Phonetic Guide (ruby)
- Options: ignore English/digits and punctuation
- Export: Word (.docx) or PDF (.pdf) using Microsoft Word (COM)
Install on Python 3.7.8 (Windows):
pip install wxPython==4.1.1 pypinyin==0.49.0 pdfminer.six==20220524 pywin32==305 python-docx==0.8.11 opencc-python-reimplemented==0.1.7
"""
import os
import sys
import threading
import time
import traceback
import wx # wxPython 4.1.1
from pypinyin import pinyin, Style # 0.49.0
try:
from opencc import OpenCC # 0.1.7
_opencc = OpenCC('t2s')
except Exception:
_opencc = None
try:
import docx # python-docx 0.8.11
except Exception:
docx = None
try:
from pdfminer.high_level import extract_text as pdf_extract_text # pdfminer.six 20220524
except Exception:
pdf_extract_text = None
try:
import win32com.client as win32 # pywin32 305
except Exception:
win32 = None
def is_cjk_char(ch: str) -> bool:
code = ord(ch)
return (
0x3400 <= code <= 0x9FFF or
0xF900 <= code <= 0xFAFF or
0x20000 <= code <= 0x2FA1F
)
def is_xml_compatible_char(ch: str) -> bool:
"""检查字符是否与XML兼容"""
code = ord(ch)
# 允许的字符范围：
# 0x20-0xD7FF: 基本可打印字符和各种语言字符
# 0xE000-0xFFFD: 私有用途和其他字符
# 0x10000-0x10FFFF: 补充字符
# 排除控制字符 (0x00-0x1F 除了 \t \n \r) 和 0xFFFE, 0xFFFF
if ch in ('\t', '\n', '\r'):
return True
if code < 0x20:
return False
if 0x20 <= code <= 0xD7FF:
return True
if 0xE000 <= code <= 0xFFFD:
return True
if 0x10000 <= code <= 0x10FFFF:
return True
return False
def clean_text_for_xml(text: str) -> str:
"""清理文本中的不兼容XML字符"""
if not isinstance(text, str):
text = str(text)
# 使用列表推导式过滤掉不兼容的字符
cleaned = ''.join(ch for ch in text if is_xml_compatible_char(ch))
# 清理多个连续的空行
cleaned = cleaned.replace('\r\n', '\n').replace('\r', '\n')
lines = cleaned.split('\n')
# 移除过多的连续空行（最多保留2个）
result_lines = []
empty_count = 0
for line in lines:
if not line.strip():
empty_count += 1
if empty_count <= 2:
result_lines.append(line)
else:
empty_count = 0
result_lines.append(line)
return '\n'.join(result_lines)
def generate_pinyin_for_text(text: str):
if not isinstance(text, str):
text = str(text)
# 在处理前先清理不兼容的字符
text = clean_text_for_xml(text)
annotated = []
for ch in text:
if is_cjk_char(ch):
# 对单个字符转换，避免转换后字符数改变导致索引错位
simp_ch = _opencc.convert(ch) if _opencc is not None else ch
pys = pinyin(simp_ch, style=Style.TONE, strict=False, heteronym=False)
py_txt = pys[0][0].strip() if pys and pys[0] else ''
annotated.append((ch, py_txt or None))
else:
# 非中文字符保持原样，不标注拼音
annotated.append((ch, None))
return annotated
def read_file_text(path: str) -> str:
ext = os.path.splitext(path)[1].lower()
if ext == '.txt':
with open(path, 'rb') as f:
data = f.read()
# 尝试多种编码
for encoding in ['utf-8', 'gbk', 'gb2312', 'utf-16']:
try:
return data.decode(encoding)
except Exception:
pass
# 如果都失败，使用 utf-8 with ignore
return data.decode('utf-8', 'ignore')
if ext == '.docx':
raise RuntimeError('暂不支持 .docx，请使用 .txt 或 .pdf')
if ext == '.pdf' and pdf_extract_text is not None:
return pdf_extract_text(path)
if ext == '.pdf' and pdf_extract_text is None:
raise RuntimeError('pdfminer.six not installed; cannot read PDF')
raise RuntimeError('Unsupported file type: %s' % ext)
def export_with_word_ruby(output_path: str, text: str, export_pdf: bool, log=None) -> str:
if docx is None:
raise RuntimeError('python-docx not available')
def _log(msg: str):
if log:
log(msg)
try:
# 创建新文档
doc = docx.Document()
_log('文本长度: %d 字符' % len(text))
# 逐行处理
normalized = text.replace('\r\n', '\n').replace('\r', '\n')
lines = normalized.split('\n')
applied = 0
_log('总行数: %d' % len(lines))
for idx, line in enumerate(lines):
if not line.strip():
continue
annotated_line = generate_pinyin_for_text(line)
chars = [ch for ch, _ in annotated_line]
pys = [(py if (py and is_cjk_char(ch)) else '') for ch, py in annotated_line]
if not chars:
continue
# 调试日志
cjk_count = sum(1 for ch in chars if is_cjk_char(ch))
_log('第 %d 行: 总字符数=%d, CJK字符数=%d' % (idx + 1, len(chars), cjk_count))
# 按行宽限制分段处理（每行最多25个字符）
max_chars_per_line = 20
line_start = 0
while line_start < len(chars):
line_end = min(line_start + max_chars_per_line, len(chars))
# 构建拼音行和字符行
pinyin_parts = []
chars_parts = []
for i in range(line_start, line_end):
ch = chars[i]
py = pys[i]
py_padded = (py if py else '').ljust(max(len(py) if py else 0, 1))
ch_padded = ch
pinyin_parts.append(py_padded)
chars_parts.append(ch_padded)
# 构建完整的行
pinyin_line = ' '.join(pinyin_parts) # 用空格分隔
chars_line = ' '.join(chars_parts)
# 再次清理以确保XML兼容性
pinyin_line = clean_text_for_xml(pinyin_line)
chars_line = clean_text_for_xml(chars_line)
# 添加到文档（使用等宽字体）
# 拼音行
p1 = doc.add_paragraph()
r1 = p1.add_run(pinyin_line)
r1.font.name = 'Courier New'
r1.font.size = docx.shared.Pt(8)
p1.paragraph_format.space_before = docx.shared.Pt(0)
p1.paragraph_format.space_after = docx.shared.Pt(0)
p1.paragraph_format.line_spacing = 1.0
# 字符行
p2 = doc.add_paragraph()
r2 = p2.add_run(chars_line)
r2.font.name = 'Courier New'
r2.font.size = docx.shared.Pt(11)
p2.paragraph_format.space_before = docx.shared.Pt(0)
p2.paragraph_format.space_after = docx.shared.Pt(0) # 行间距
p2.paragraph_format.line_spacing = 1.0
applied += sum(1 for i in range(line_start, line_end) if is_cjk_char(chars[i]))
line_start = line_end
# 行间空行
if idx < len(lines) - 1:
p_empty = doc.add_paragraph()
p_empty.paragraph_format.space_before = docx.shared.Pt(0)
p_empty.paragraph_format.space_after = docx.shared.Pt(0)
_log('拼音标注完成，共标注字符: %d' % applied)
# 保存文档
docx_path = output_path if output_path.lower().endswith('.docx') else (output_path + '.docx')
doc.save(docx_path)
_log('已导出Word: %s' % docx_path)
# 如果需要 PDF
if export_pdf:
pdf_path = output_path if output_path.lower().endswith('.pdf') else (output_path + '.pdf')
if win32 is not None:
try:
_log('转换PDF中...')
word = win32.Dispatch('Word.Application')
try:
word.Visible = False
word_doc = word.Documents.Open(docx_path)
word_doc.SaveAs2(pdf_path, FileFormat=17)
word_doc.Close(False)
_log('已导出PDF: %s' % pdf_path)
return pdf_path
finally:
word.Quit()
except Exception as e:
_log('PDF转换失败: %s，保留DOCX文件' % str(e))
return docx_path
else:
_log('pywin32 不可用，无法转换PDF')
return docx_path
return docx_path
except Exception as e:
_log('导出失败: %s' % str(e))
_log(traceback.format_exc())
raise
class MainFrame(wx.Frame):
def __init__(self):
super(MainFrame, self).__init__(parent=None, title='文档拼音注释工具', size=(920, 640))
self.CenterOnScreen()
panel = wx.Panel(self)
vbox = wx.BoxSizer(wx.VERTICAL)
hbox1 = wx.BoxSizer(wx.HORIZONTAL)
self.btn_files = wx.Button(panel, label='选择文件(txt/pdf)')
self.btn_files.Bind(wx.EVT_BUTTON, self.on_pick_files)
hbox1.Add(self.btn_files, 0, wx.RIGHT, 8)
self.btn_clear = wx.Button(panel, label='清空列表')
self.btn_clear.Bind(wx.EVT_BUTTON, self.on_clear_list)
hbox1.Add(self.btn_clear, 0)
vbox.Add(hbox1, 0, wx.ALL, 10)
self.list_files = wx.ListBox(panel, style=wx.LB_EXTENDED)
vbox.Add(self.list_files, 1, wx.EXPAND | wx.LEFT | wx.RIGHT, 10)
vbox.Add(wx.StaticText(panel, label='或直接在下方粘贴文本：'), 0, wx.LEFT | wx.RIGHT | wx.TOP, 10)
self.text_input = wx.TextCtrl(panel, style=wx.TE_MULTILINE)
vbox.Add(self.text_input, 1, wx.EXPAND | wx.ALL, 10)
opt_box = wx.BoxSizer(wx.HORIZONTAL)
self.choice_format = wx.Choice(panel, choices=['导出为Word(.docx)', '导出为PDF(.pdf)'])
self.choice_format.SetSelection(0)
opt_box.Add(self.choice_format, 0, wx.RIGHT, 15)
self.btn_outdir = wx.Button(panel, label='选择输出目录')
self.btn_outdir.Bind(wx.EVT_BUTTON, self.on_pick_outdir)
opt_box.Add(self.btn_outdir, 0)
vbox.Add(opt_box, 0, wx.LEFT | wx.RIGHT | wx.BOTTOM, 10)
# 限制提示
self.max_file_mb = 20 # 最大文件大小（MB）
self.max_paste_chars = 200000 # 粘贴文本最大字符数
vbox.Add(wx.StaticText(panel, label='提示：最大支持文件大小 %d MB，粘贴文本最大 %d 字符' % (
self.max_file_mb, self.max_paste_chars)), 0, wx.LEFT | wx.RIGHT | wx.BOTTOM, 10)
self.btn_run = wx.Button(panel, label='开始处理')
self.btn_run.Bind(wx.EVT_BUTTON, self.on_run)
vbox.Add(self.btn_run, 0, wx.ALL, 10)
vbox.Add(wx.StaticText(panel, label='处理日志：'), 0, wx.LEFT | wx.RIGHT, 10)
self.log = wx.TextCtrl(panel, style=wx.TE_MULTILINE | wx.TE_READONLY)
vbox.Add(self.log, 1, wx.EXPAND | wx.ALL, 10)
panel.SetSizer(vbox)
self.files = []
# 默认保存目录：桌面
self.outdir = os.path.join(os.path.expanduser('~'), 'Desktop')
if not os.path.exists(self.outdir):
os.makedirs(self.outdir)
def append_log(self, msg: str):
ts = time.strftime('%H:%M:%S')
self.log.AppendText('[%s] %s\n' % (ts, msg))
def on_pick_files(self, event):
dlg = wx.FileDialog(self, message='选择文件', wildcard='文档|*.txt;*.pdf',
style=wx.FD_OPEN | wx.FD_FILE_MUST_EXIST | wx.FD_MULTIPLE)
if dlg.ShowModal() == wx.ID_OK:
paths = dlg.GetPaths()
self.files = list(paths)
self.list_files.Clear()
for p in self.files:
self.list_files.Append(p)
dlg.Destroy()
def on_clear_list(self, event):
self.files = []
self.list_files.Clear()
def on_pick_outdir(self, event):
dlg = wx.DirDialog(self, message='选择输出目录', style=wx.DD_DEFAULT_STYLE | wx.DD_DIR_MUST_EXIST)
if dlg.ShowModal() == wx.ID_OK:
self.outdir = dlg.GetPath()
self.append_log('输出目录: %s' % self.outdir)
dlg.Destroy()
def on_run(self, event):
export_pdf = self.choice_format.GetSelection() == 1
if not self.files and not self.text_input.GetValue().strip():
wx.MessageBox('请选择文件或粘贴文本', '提示', wx.OK | wx.ICON_WARNING)
return
self.btn_run.Disable()
self.log.Clear()
def worker():
try:
pasted = self.text_input.GetValue().strip()
if pasted and len(pasted) > self.max_paste_chars:
raise RuntimeError('粘贴文本超出最大限制（%d 字符）' % self.max_paste_chars)
if pasted:
name = 'pasted_%s' % time.strftime('%Y%m%d_%H%M%S')
out_path = os.path.join(self.outdir, name)
self.append_log('处理粘贴文本...')
export_with_word_ruby(out_path, pasted, export_pdf, self.append_log)
for idx, f in enumerate(self.files, 1):
self.append_log('[%d/%d] 处理 %s' % (idx, len(self.files), f))
try:
# 文件大小限制
try:
size_mb = os.path.getsize(f) / (1024.0 * 1024.0)
if size_mb > self.max_file_mb:
raise RuntimeError('文件超过大小限制（%.2f MB > %d MB）' % (size_mb, self.max_file_mb))
except Exception:
pass
text = read_file_text(f)
base = os.path.splitext(os.path.basename(f))[0]
out_path = os.path.join(self.outdir, base)
export_with_word_ruby(out_path, text, export_pdf, self.append_log)
except Exception:
self.append_log('处理失败: %s' % f)
self.append_log(traceback.format_exc())
wx.CallAfter(wx.MessageBox, '处理完成', '提示', wx.OK | wx.ICON_INFORMATION)
except Exception as e:
self.append_log('发生错误: %s' % str(e))
self.append_log(traceback.format_exc())
wx.CallAfter(wx.MessageBox, str(e), '错误', wx.OK | wx.ICON_ERROR)
finally:
wx.CallAfter(self.btn_run.Enable)
t = threading.Thread(target=worker, daemon=True)
t.start()
def main():
app = wx.App(False)
frame = MainFrame()
frame.Show()
app.MainLoop()
if __name__ == '__main__':
main()