---
license: gpl-3.0
language:
- zh
pipeline_tag: audio-classification
---
# Voiceprint Recognition model of zhvoice based on EcapaTdnn
This is a VPR model trained on [**zhvoice**](https://aistudio.baidu.com/aistudio/datasetdetail/133922) dataset using [**yeyupiaoling/VoiceprintRecognition-Pytorch**](https://github.com/yeyupiaoling/VoiceprintRecognition-Pytorch). I choosed MelSpectrogram as preprocessing method.

You can use it with my [**2DIPW/audio_dataset_vpr**](https://github.com/2DIPW/audio_dataset_vpr) project (modified from yyupiaoling/VoiceprintRecognition-Pytorch), or the origin project.