Liujgoj-Cantonese-Qwen2.5-Omni-7B-SFT-V2.2
📌 Model Overview
Liujgoj-Cantonese-Qwen2.5-Omni-7B-SFT-V2.2 is the Supervised Fine-Tuned (SFT) version of Liujgoj-Cantonese-Qwen2.5-Omni-7B-CPT-V2. It is designed to follow explicit instructions in Cantonese NLP and speech-related pipelines, with a focus on Liujgoj Cantonese orthography, romanization, syllable decomposition, and IPA transcription.
While the CPT stage taught the model the statistical structure and phonological distribution of Liujgoj Cantonese, this SFT stage teaches it to follow user instructions and perform discrete conversion tasks with high accuracy.
- Developer: Yvthyvq / Liujgoj Cantonese Project
- Base Model: Yvthyvq/Liujgoj-Cantonese-Qwen2.5-Omni-7B-CPT-V2
- Model Type: Instruction-Tuned / Supervised Fine-Tuned (SFT)
- Primary Language/Script: Cantonese (Liujgoj Latin Orthography & Traditional Chinese Characters)
- Model Size: 7B (full Omni architecture preserved)
🏗️ Model Architecture Note
This checkpoint is a complete Qwen2.5-Omni model, not a text-only derivative. The SFT LoRA adapter was merged back into the thinker submodule, while the original talker and token2wav components were preserved intact.
This design enables the following pipeline: Qwen2.5-Omni-7B (base) │ ▼ [CPT] Liujgoj Cantonese text continuation │ ▼ [SFT] Instruction tuning (this checkpoint) │ ▼ [ASR] Cantonese speech recognition fine-tuning (upcoming)
Because the full multimodal architecture is retained, this checkpoint can be directly used as the starting point for downstream speech tasks (ASR / TTS) without re-loading the original Omni weights.
🛠️ Training Details & Methodology
訓練命令：train_lora_sft.py
訓練過程：train_sft.txt
基本測試：
【測試 1】 Prompt: 請將以下粵語漢字轉寫為羅馬字： 喺江南靠近海邊一條大路之上，有一隊清兵，手執刀槍，押住七架囚車，向住北方行緊。
--- 模型輸出 --- Hair Gongjnaamx kaauganh hoirminh yatj tiux daaihlouh zijsoengh, yauq yatj deoih cingjbingj, saur zapj doujcoengj, aat zvh catj gaa cauxcej, hoeng zvh bakjfungj haangxganr.
【測試 2】 Prompt: 請將以下羅馬字轉寫為粵語漢字： Waahsvt hair Cingjciux ge Hongjheij cojninx, dungjtinj ge mauq yatj yath. batjfungj yvx douj, munq deih bingjsoengj.
--- 模型輸出 --- 話說喺清朝嘅康熙初年，冬天嘅某一日。北風如刀，滿地冰霜。
【測試 3】 Prompt: 請將「waahsvt」拆分成音節，並標註音標。
--- 模型輸出 --- waah + svt
大規模測試結果見：水滸傳test.xlsx
註： 第一列係原來嘅漢字，第二列爲轉寫結果。
數據來源：CanCLID/zoengjyutgaai（https://huggingface.co/datasets/CanCLID/zoengjyutgaai）。
測試命令：transcribe_xlsx.py。
🤝 Dataset Acknowledgements & Licenses
Yvthyvq/ZoengjYvthgaaij-MouxZaakhdungj-Liujgoj
Yvthyvq/HKCanCor-Liujgoj
Multi-film Cantonese dialogue corpus (cleaned and transliterated into Liujgoj orthography)
We thank the original authors and maintainers of these corpora for making Cantonese NLP research possible.
📝 Changelog
v2.2 (SFT release): Instruction tuning on 150,846 multi-task samples covering romanization, syllable decomposition, and IPA transcription. Full Omni architecture preserved for downstream ASR.