香港圍頭話及客家話文字轉語音

_{Hong Kong Waitau & Hakka Text-to-Speech}

本儲存庫包含香港圍頭話及客家話文字轉語音之預生成音訊檔案。當選擇輕巧模式時，朗讀器程式會從中選取音訊片段拼接，於程式內直接產生音訊。

This repository contains the pre-generated audio files of the Hong Kong Waitau & Hakka Text-to-Speech reader. In lightweight mode, the TTS application generates audio by concatenating selected audio segments right inside the app.

These files are created as follows:

• For each language (waitau and hakka):
  • For each voice (male and female):
    • For each character in the dictionary, an audio file is generated using the model for offline inference.
    • The generated files are then concatenated into a single file, chars.bin, and the corresponding pronunciation and the start offset for each audio file is saved into an offset table, chars.csv.
    • The same is performed for each word in the dictionary, producing the files words.bin (split into smaller chunks due to size limitations) and words.csv.

For how they are used by the main app, refer to the Audio Generation section in the README of the TTS repo.