---
license: apache-2.0
language:
  - zh
tags:
  - speaker-verification
  - speaker-recognition
  - speaker-embedding
  - speech
  - audio
  - campplus
pipeline_tag: audio-classification
---

# CAMPPlus Speaker Encoder

This directory contains a local copy of the CAMPPlus speaker verification model, originally released through ModelScope / 3D-Speaker. The model is used to extract speaker embeddings from speech, while the original model architecture, weights, authorship, and license remain attributed to the original authors.

The upstream 3D-Speaker project is released under the `Apache-2.0` license. When using, redistributing, or citing these files, please keep the original author information, model link, and license notice.

Original links:

- ModelScope: https://www.modelscope.cn/models/iic/speech_campplus_sv_zh-cn_16k-common
- 3D-Speaker: https://github.com/modelscope/3D-Speaker
- License: https://www.apache.org/licenses/LICENSE-2.0