G2PWModel_zh-Hans

中文 | English

---

中文

简介

这是 G2PW (Grapheme-to-Phoneme for Word) 模型的重新训练版本，专门针对简体中文进行了优化。

主要优化

• 纯简体中文数据集：此版本仅使用简体中文数据进行训练，提供更准确的简体中文发音预测
• 更快的推理速度：推荐使用 @baicai-1145/g2pw-torch 进行推理，速度比原版 ONNX 实现更快

模型信息

• 模型名称: G2PWModel_zh-Hans
• 训练数据: 简体中文语料
• 优化目标: 提高简体中文字音转换准确率

使用方法

推荐使用 g2pw-torch 进行推理：

pip install g2pw-torch

from g2pw import G2PWConverter

converter = G2PWConverter(model_dir='baicai1145/G2PWModel_zh-Hans')
result = converter('我今天很开心')
print(result)

致谢

• 数据收集与整理: @TheSmallHanCat
• 模型训练: @baicai1145

相关项目

• baicai-1145/g2pW - 原始 G2PW 项目
• baicai-1145/g2pw-torch - PyTorch 推理版本（推荐使用）

许可证

本项目遵循原项目的许可证协议。

---

English

Introduction

This is a retrained version of the G2PW (Grapheme-to-Phoneme for Word) model, specifically optimized for Simplified Chinese.

Key Improvements

• Simplified Chinese Only Dataset: This version is trained exclusively on Simplified Chinese data, providing more accurate pronunciation predictions for Simplified Chinese
• Faster Inference Speed: We recommend using @baicai-1145/g2pw-torch for inference, which is faster than the original ONNX implementation

Model Information

• Model Name: G2PWModel_zh-Hans
• Training Data: Simplified Chinese corpus
• Optimization Goal: Improve accuracy of Simplified Chinese grapheme-to-phoneme conversion

Usage

We recommend using g2pw-torch for inference:

pip install g2pw-torch

from g2pw import G2PWConverter

converter = G2PWConverter(model_dir='baicai1145/G2PWModel_zh-Hans')
result = converter('我今天很开心')
print(result)

Credits

• Data Collection and Organization: @TheSmallHanCat
• Model Training: @baicai1145

Related Projects

• baicai-1145/g2pW - Original G2PW project
• baicai-1145/g2pw-torch - PyTorch inference version (recommended)

License

This project follows the same license as the original project.