hispark-modelzoo-fsmn-kws

FSMN-KWS

小云唤醒词检测模型基于CharCTC（Character-level Connectionist Temporal Classification）架构，用于检测语音中的特定唤醒词"小云小云"。模型提取音频的Fbank特征，通过LFR（Low Frame Rate）降采样和CMVN归一化后进行CTC解码，输出关键词检测结果。

Mirror Metadata

• Hugging Face repo: shadow-cann/hispark-modelzoo-fsmn-kws
• Portal model id: ku4ntpbgj400
• Created at: 2026-09-14 09:53:03
• Updated at: 2026-09-15 15:27:10
• Category: 视频

Framework

• PyTorch

Supported OS

• Linux

Computing Power

• Hi3516CV610

Tags

• 关键字检测

Detail Parameters

• 计算量: 0.229GFLOPs
• 输入: 151x400
• 参数量: 0.758M

Files In This Repo

• speech_charctc_kws_phone-xiaoyun_onnxsim_364.onnx (源模型 / 源模型元数据)
• speech_charctc_kws_phone-xiaoyun_onnxsim_151.onnx (源模型 / 源模型元数据)
• speech_charctc_kws_phone-xiaoyun_364.onnx (源模型 / 源模型元数据)
• speech_charctc_kws_phone-xiaoyun_151.onnx (源模型 / 源模型元数据)
• kws-610.om (编译模型 / OM 元数据 / A8W8)
• kws-610-a.om (编译模型 / OM 元数据 / A8W8)

Upstream Links

• Portal card: https://gitbubble.github.io/hisilicon-developer-portal-mirror/model-detail.html?id=ku4ntpbgj400
• Upstream repository: https://gitcode.com/HiSpark/modelzoo/tree/master/samples/built-in/audio/fsmn_kws
• License reference: https://www.modelscope.cn/iic/speech_charctc_kws_phone-xiaoyun.git

Notes

• This repository was mirrored from the HiSilicon Developer Portal model card and local downloads captured on 2026-03-27.
• File ownership follows the portal card mapping, not just filename similarity.
• Cover image: 1763392620658690___.JPG