| funasr/models/e2e_asr_mfcca.py | ●●●●● 补丁 | 查看 | 原始文档 | blame | 历史 |
funasr/models/e2e_asr_mfcca.py
@@ -36,7 +36,11 @@ import random import math class MFCCA(AbsESPnetModel): """CTC-attention hybrid Encoder-Decoder model""" """ Author: Audio, Speech and Language Processing Group (ASLP@NPU), Northwestern Polytechnical University MFCCA:Multi-Frame Cross-Channel attention for multi-speaker ASR in Multi-party meeting scenario https://arxiv.org/abs/2210.05265 """ def __init__( self,