Instructions to use mesut/turkish_speech_captioner_llama-3.1_lora_model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mesut/turkish_speech_captioner_llama-3.1_lora_model with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("mesut/turkish_speech_captioner_llama-3.1_lora_model", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Unsloth Desktop
Uploaded model
The unsloth/meta-llama-3.1-8b-bnb-4bit model is fine tuned by 30.000 mp speeches and captions in Turkish. The data is curated from Parliament web page. Sytem promt is below.
system_prompt = """Below is an instruction that describes a speech in Turkish. Write a caption as response that appropriately completes the request. All data in Turkish and response should be in Turkish as well.
Instruction:
Konuşma Metni: {speech}
Input:
{''}
Response:
Konuşma Başlığı: {caption}"""
- Developed by: mesut
- License: apache-2.0
- Finetuned from model : unsloth/meta-llama-3.1-8b-bnb-4bit
This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
