I’m looking for a free tool where I can input any text and get one of these tones in return: neutral, formal, humorous, romantic, or attractive. Can you help me?
I tried to use IBM Watson, but I wasn’t successful. I also asked ChatGPT to detect tone for each sentence, which gave me good feedback, but I don’t want to use LLM and seeking for specific model or a light fine-tuned language model for this task.
Maybe Text-classification models?
Hi, I recently participated in a hackathon where the problem was identifying the tonality of blog text and seeing if they matched with organization guidelines. During the research, I was not able to find a ML model. I tried to generate synthetic data (since I couldn’t find any public dataset available) for each tonality category using LLM. Then, based on the data, I tried to train a ML model (SVM, XGBoost with Hyperparameters tuned). However the results were not good.
Finally, I resorted to utilizing LLM for the classification and then built on the other requirements of the hackathon. So a summary response to your question: based on my research and limited knowledge: you might have to go the long route of creating a synthetic data set and an ML model which might not be so accurate due to the size of the training data set. Or go towards the LLM route which I have observed to always yield better result. Hope this helps.