---
title: speaker_recognition_task_models_onnx_collection
canonical_url: "https://www.modelscope.cn/models/manyeyes/speaker_recognition_task_models_onnx_collection"
md_url: "https://www.modelscope.cn/models/manyeyes/speaker_recognition_task_models_onnx_collection.md"
repository: manyeyes/speaker_recognition_task_models_onnx_collection
chinese_name: "说话人识别（说话人确认）onnx模型合集"
last_updated: 2024-02-07
pipeline_tag: speaker-verification
tasks:
  - speaker-verification
library_name:
  - onnx
  - "其他"
frameworks:
  - "其他"
downloads: 978
stars: 10
---

# speaker_recognition_task_models_onnx_collection

> speaker_recognition_task_models_onnx_collection - manyeyes 在 ModelScope 开源的模型。从wespeaker，3d-speaker项目导出的说话人识别（说话人确认）的 onnx 模型。模型名称结构：项目名+模型名。

- **Repository**: manyeyes/speaker_recognition_task_models_onnx_collection
- **Tasks**: speaker-verification
- **Downloads**: 978
- **Stars**: 10
- **Last updated**: 2024-02-07

Source: https://www.modelscope.cn/models/manyeyes/speaker_recognition_task_models_onnx_collection

---

###### 当前模型用于说话人识别（说话人确认）。
###### 模型是从wespeaker，3dspeaker中导出的onnx模型。
#### 输入和输出参数

##### INPUTS

name: feats

tensor: float32[B,T,80]

##### OUTPUTS

name: embs

tensor: float32[B,*] 

*的值根据具体模型会有所不同，比如256，192 …… ^_^

#### Clone with HTTP
```bash
 git clone https://www.modelscope.cn/manyeyes/speaker_recognition_task_models_onnx_collection.git
```