Lyric Learn

Japanese Lyric Learner is a Python application designed to analyze Japanese song lyrics. It uses the OpenAI API to provide detailed translations and explanations of the lyrics.

Features

• Analyzes Japanese song lyrics in batches.
• Provides accurate translations and detailed explanations of grammar and key vocabulary.
• Saves the analysis results to output files.

Requirements

• Python 3.11+
• uv package manager (recommended) or pip
• Required Python packages (listed in pyproject.toml)

Installation

Using uv (recommended)

1. Clone the repository:

   git clone <repository-url>
   cd japanese-lyric-learner

2. Install uv if you haven't already:

   curl -LsSf https://astral.sh/uv/install.sh | sh

3. Install dependencies:

   uv sync

4. Create a .env file in the root directory and add your OpenAI API key, model name, and base URL:

   OPENAI_API_KEY=your_openai_api_key
   MODEL_NAME=qwen-plus
   BASE_URL=https://dashscope-intl.aliyuncs.com/compatible-mode/v1

5. If you want to use the OpenAI GPT model, update your .env file as follows:

   OPENAI_API_KEY=your_openai_api_key
   MODEL_NAME=gpt-5
   BASE_URL=https://api.openai.com/v1  # Default OpenAI API URL

Usage

1. Place your song lyrics in the original_song/learned or original_song/to_learn directories.

2. Run the main script:

   uv run main.py

This will create a file named test_analyze.txt in the output directory, which you can use as a reference for the analysis results.