manotham/Thai-dialogue-translate_emotion_mdpov2_ckp269
The manotham/Thai-dialogue-translate_emotion_mdpov2_ckp269 is a 4 billion parameter Qwen3-based model fine-tuned by manotham for English to Thai dialogue translation. Optimized with Unsloth and DPO, its primary differentiator is emotion-controlled translation, allowing users to specify emotional tones like joy, sadness, or anger. This model excels at translating cinematic, game, and visual novel dialogue while preserving the intended emotional nuance.
Loading preview...
Thai Dialogue Emotion-Aware Translator
This model, developed by manotham, is a Qwen3-based language model specifically fine-tuned for English to Thai dialogue translation. It leverages Unsloth for efficient training and Direct Preference Optimization (DPO) for enhanced naturalness and emotion alignment in translations.
Key Capabilities
- Emotion-Controlled Translation: Translates dialogue while preserving specified emotional tones such as sadness, anger, contempt, joy, gratitude, and more, based on the
tabularisai/multilingual-emotion-classificationframework. - Natural Thai Dialogue: Focuses on generating natural-sounding Thai suitable for conversational contexts.
- Optimized for Low VRAM: Designed to be quantization-friendly and tested on consumer GPUs.
Ideal Use Cases
This model is particularly well-suited for applications requiring emotionally nuanced dialogue translation, including:
- Games and Visual Novels
- Cinematic Dialogue
- Storytelling and Fantasy/Anime-style Conversations
Limitations
It is important to note that the model is optimized for dialogue and may not perform optimally for formal documents or highly technical content. Some outputs might exhibit an overly dramatic tone, and there's a possibility of overfitting to fantasy/anime dialogue styles.