-
https://github.com/TaoRuijie/ECAPA-TDNN
-
## error log | 日志或报错信息 | ログ
## model | 模型 | モデル
1. original model
model.trace.pt ,是声纹识别中的ecapa_tdnn模型
## how to reproduce | 复现步骤 | 再現方法
转换之后,pnnx和ncnn的输出不一致,ncnn 和 pnnx版本是 20240410
```
python…
-
您好感谢你们公开的工作,不过有个问题想请教下。我在尝试examples/aishell3/vc2 TCAPA-TDNN的voice clone时,我直接跑了run.sh中voice_clone这部。我使用的测试src是我自己的原声(男声),但是tts出来的结果竟然是女声。我使用的都是预训练模型。请问这是不是spk_embedding有问题也就是tcapa-tdnn提取的有问题。
-
### Description
Train ECAPA-TDNN on adults' and children's voices
### Tasks
- [ ] explore the impact of training data window on speaker diarization performance (in a child-adult setting) and on spe…
-
According to the WavLM paper:
([WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://arxiv.org/pdf/2110.13900.pdf))
They used ECAPA-TDNN embeddings model for …
-
When the first epoch ends, I've an error in the evaluation because there are some speakers/labels that are not present on label_encoder.txt
```
(.venv) root@8ad8297faf3e:/home/diarization/speechbr…
-
Thank you for providing the eval script.
I found a hard-coding path `/mnt/bn/jdy-lq-2/s3prl_s3prl_main` in `thirdparty/UniSpeech/downstreams/speaker_verification/models/ecapa_tdnn.py`, which caused …
-
I donnot find the ecapa-tdnn code, is there noly the code of tdnn?
-
Thanks for your work. In the paper, you mentioned that the teacher model is wavLM + ECAPA-TDNN. However, in your implementation, I only found you loaded the weights of wavLM (line 13 in train/experime…
gancx updated
3 months ago
-
- The ECAPA-TDNN model trained on "development part of the `VoxCeleb2`".
- The datasets contains 5994 speakers.
- Some speakers have many more files than other speakers (`imbalanced classes)`.
- …