---
title: latin_whisper-small
canonical_url: "https://www.modelscope.cn/models/AI-ModelScope/latin_whisper-small"
md_url: "https://www.modelscope.cn/models/AI-ModelScope/latin_whisper-small.md"
repository: AI-ModelScope/latin_whisper-small
last_updated: 2025-10-13
license: cc-by-4.0
model_type:
  - whisper
architectures:
  - WhisperForConditionalGeneration
base_model:
  - microsoft/speecht5_tts
base_model_relation: finetune
parameters: 241.7M
tensor_type:
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - la
inference_backends:
  - "deploy_task vlm/text/emb"
  - "vllm 0.9.2"
downloads: 102
stars: 0
---

# latin_whisper-small

> latin_whisper-small - AI-ModelScope 在 ModelScope 开源的模型。Whisper Small Latin - Ken This model is a fine-tuned version of openai/whisper-small on the Latin language. It achieves a 20 CER (with punctuations) with the 67 hours of audio from Vox Classica

AI-ModelScope/latin_whisper-small 是 ModelScope 魔搭社区上的 241.7M 参数机器学习模型，采用 cc-by-4.0 许可，基于 microsoft/speecht5_tts 构建，可用 deploy_task vlm/text/emb、vllm 0.9.2 部署。

- **Repository**: AI-ModelScope/latin_whisper-small
- **License**: cc-by-4.0
- **Parameters**: 241.7M
- **Base model**: microsoft/speecht5_tts
- **Inference backends**: deploy_task vlm/text/emb, vllm 0.9.2
- **Downloads**: 102
- **Stars**: 0
- **Last updated**: 2025-10-13

Source: https://www.modelscope.cn/models/AI-ModelScope/latin_whisper-small

---

# Whisper Small Latin - Ken
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Latin language. It achieves a 20 CER (with punctuations) with the 67 hours of audio from [Vox Classica](https://huggingface.co/datasets/Ken-Z/Latin-Audio)

Try this model directly in your browser: [ASR Demo](https://huggingface.co/spaces/ken-z/latin_whisper-small-demo)