Issues

• Question about pretrained BSRNN Online checkpoint (avg_best_model.pt)

  wenet-e2e/wesep#38 · Filter by author rahulp-coder opened on Jul 9, 2026
  1

• Idea use arcface wakeword for wakeword and embeddings.

  wenet-e2e/wesep#37 · Filter by author rolyantrauts opened on Apr 19, 2026

• 有详细的使用文档吗？

  wenet-e2e/wesep#36 · Filter by author liujian19911023 opened on Mar 5, 2026

• bug in bsrnn_feats.py

  wenet-e2e/wesep#35 · Filter by author haha010508 opened on Nov 18, 2025
  8

• Usage for TSE for a particular speaker

  wenet-e2e/wesep#32 · Filter by author mukherjeesougata-eros opened on Sep 29, 2025
  4

• ONNX export of BSRNN model

  wenet-e2e/wesep#29 · Filter by author sangeet2020 opened on Mar 26, 2025
  1

• real time inference

  wenet-e2e/wesep#27 · Filter by author heartQi opened on Mar 7, 2025

• BSRNN model initialization is failing for 8kHz sampling rate

  wenet-e2e/wesep#25 · Filter by author sriyachakravarthy opened on Feb 27, 2025
  1

• To achieve the best performance

  wenet-e2e/wesep#24 · Filter by author flect-mori6638 opened on Feb 25, 2025
  2

• For Non-GPU users

  wenet-e2e/wesep#23 · Filter by author jjy714 opened on Feb 17, 2025

• What if there is only one speaker in mix audio, and he/she is not the enrollment person

  wenet-e2e/wesep#22 · Filter by author SherryYu33 opened on Feb 8, 2025
  11

• The issue of extracting a different speaker than expected.

  wenet-e2e/wesep#20 · Filter by author flect-mori6638 opened on Feb 4, 2025
  5

• Train Loss is NAN after 13 epoch

  wenet-e2e/wesep#18 · Filter by author heartQi opened on Jan 13, 2025
  9

• problem about dataloader/num_workers

  wenet-e2e/wesep#17 · Filter by author Shookwu opened on Dec 10, 2024
  3

• Pretrained model

  wenet-e2e/wesep#3 · Filter by author yonzhao opened on Sep 5, 2024
  9