Instructions to use MigoXV/dnsmos-p835 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MigoXV/dnsmos-p835 with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("MigoXV/dnsmos-p835", device_map="auto") - Notebooks
- Google Colab
- Kaggle
DNSMOS P.835
DNSMOS P.835 的 Hugging Face Safetensors checkpoint。模型接收 16 kHz、9.01 秒单声道波形,输入形状为 [batch, 144160],输出 SIG、BAK、OVR 三个原始分数。
模型架构由 mos_assess 注册到 Transformers AutoClass:
import torch
from transformers import AutoModel
import mos_assess
model = AutoModel.from_pretrained("MigoXV/dnsmos-p835")
waveform = torch.zeros(1, model.config.input_num_samples)
with torch.inference_mode():
scores = model(input_values=waveform).scores
权重文件使用 Safetensors 格式,不需要 trust_remote_code=True。
- Downloads last month
- 15
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support