-
D:\PaddleSpeech-develop\demos\speech_web\speech_server>python main.py --port 8010
2022-10-11 03:52:59.439 | INFO | paddlespeech.s2t.modules.ctc::45 - paddlespeech_ctcdecoders not installed!
[nlt…
-
我发现您的这个项目中训练时,在验证的时候用的时test list,但是在eval的时候也是用的test去测试模型的效果,是不是在train的时候验证的时候用dev集比较好
-
Dear all:
I noticed that when training on voxceleb1+2, it will take me up to 25 hours for single epoch. and even with ddp on 4 gpu cards, the training speed does not reduce at all. I guess the cpu …
-
Hello:
When I was doing the speaker differentiation on AMI corpus experiment, one parameter was unclear. cosis used to calculate the matrix similarity. What does nn mean here? Please elaborate. Thank…
-
When looking in the [speechbrain/speechbrain/pretrained/interfaces.py](https://github.com/speechbrain/speechbrain/blob/4ea7fe1e630baffb0109bb5d2ebf53d319912a62/speechbrain/pretrained/interfaces.py#L24…
-
我看到推理代码中:
with torch.no_grad():
embedding_1 = self.speaker_encoder.forward(data_1, aug = False)
embedding_1 = F.normalize(embedding_1, p=2, dim=1)
embedding_2 = self.speaker_encoder.fo…
-
Hi,
Nice job ! Thanks for sharing.
I am trying to use your code to run VoxConverse. I tried to use ground-truth VAD results for clustering, but I found that VAD results in VBx/VAD/final_syste…
-
Is there a way to match an encoding to the voxelceleb database. Analyze a voice and know which celebrity it is? Thanks
All I could see was comparing two voices with encodings.
Currently:
```
…
-
您好 我请问一下,在使用ecapa_tdnn,模型推理时候,特征提取选择spectrogram,但是得到的特征只是一个值是什么原因呢,并且对比任意两个音频相似度都是1。使用的是下面的预训练模型。
![a66545f1c2a39f8ea6527f6bd6e17b2](https://user-images.githubusercontent.com/79695576/172316183-2c661…
-
**Describe the bug**
I've fine-tuned TitaNet-Large model 10 Epoch for Korean with 1,000,000 datas and 60 speaker dataset because NGC TitaNet(for English) only predict one token. I've checked loss h…