Modelpublic

ganbon/novel-sentence-bert-base-tone-embedding

sourceHugging Faceapache-2.0updated 8mo agoView on Hugging Face

0likes76downloads

Model Card

Novel-Sentence-BERT-Base-Tone-Embedding

This model is a sentence vector model that expresses the tone of utterance in novels. We trained the model using dialogue data from 190 novels published on “小説家になろう”.

How to Use

shell

$ pip install -U sentence-transformers

Then you can use the model like this:

python

from sentence_transformers import SentenceTransformer

model_path = "ganbon/novel-sentence-bert-base-tone-embedding"
sentences = ["今日はいい天気ですわ", "この紅茶は本当においしいですわ", "今日はいい天気だな"]
model = SentenceTransformer(model_path)
embeddings = model.encode(sentences)
anchor, anchor_embedding  = sentences[0], embeddings[0]
print(f"anchoer:{anchor}")
for embedding, sentence in zip(embeddings[1:], sentences[1:]):
    print(f"{sentence}: {model.similarity(anchor_embedding, embedding).item():.2}")

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)

Citing & Authors

bib

@inproceedings{iwamoto-etal-2026-novel2dialcorpus,
    title = " Novel2DialCorpus:小説を用いた対話コーパスの自動構築手法",
    author = "岩本 和真 and 安藤 一秋",
    booktitle = "言語処理学会第32回年次大会発表論文集",
    year = "2026"
}