PyTorch_Speaker_Verification
PyTorch_Speaker_Verification copied to clipboard
HarryVolek
PyTorch implementation of "Generalized End-to-End Loss for Speaker Verification" by Wan, Li et al.
Hello , I have trained all example data, and made test . But question is : How to use speaker verification ? I have some .wav file , how can...
HI HarryVolek, I have created embeddings for Timit model and generated the d-vectors as you explained in README but when i try to test the features in USI_RNN it throw...
refactoring code with var unpack which is more pythonic, concise, readable and efficient; how do think this change which has practical value？
I run the train_speech_embedder.py with default config, but I get the result : EER 0.0723 with final checkpoint ,get the result :EER 0.0578 with intermediate checkpoint @HarryVolek @seandickert @mazzzystar @pranshurastogi29
hello，can this project be used to identify a new speaker? I mean a wav which belongs to none of the enrollment utterance. @HarryVolek
Hi Harry; I want to use d-vector for diarization with 8kHz data. I have 9000 speakers. However my loss saturate around 5 (at 250 epoch)(Should I train with more epochs?)....
Hello guys, I'm new to this tool. I would like to know the path of the stones to calculate the similarity between two audio files. I started making a code...
I used 1000 speakers, each with 100 sentences. In config.yaml, should M be set to 100? M is the number of utterances per speaker
As a results the model is trained on `N*M` utterances per epoch and not the whole training set. This affects the convergence as well as possible extensions of the code...
Hi, sir. I don't think that the perm and unperm acitions in your code make any difference. Because the perm action is along the batch dim, in the forward process,...
Metadata
Owner
Metadata
PyTorch implementation of "Generalized End-to-End Loss for Speaker Verification" by Wan, Li et al.