search

zidu07 / multimodal-biometric-verification-of-face-and-voice-using-feature-level-fusion

multimodal biometric verification of face and voice using fusion
0 stars 0 forks source link

Voice Recognition part #1

Open a1once opened 1 year ago

a1once commented 1 year ago

Hello!

I could run successfully the face recognition part. However, running your code for the voice recognition part is raising problems. Could you please share your dataset and triplet_loss_trained_model.pth file?

zidu07 commented 1 year ago

H I

The error is probably because your missing the triplet_loss_trained_model.pth file , and the data set for the voice recognition should be in .wav format.

Unfortunately the file is missing in my drive, however you can try and find that file in Google, however it might take some searching.

On Fri, 24 Feb 2023 at 8:16 pm, a1once @.***> wrote:

Hello!

I could run successfully the face recognition part. However, running your code for the voice recognition part is raising problems. Could you please share your dataset and triplet_loss_trained_model.pth file?

— Reply to this email directly, view it on GitHub https://github.com/zidu07/multimodal-biometric-verification-of-face-and-voice-using-feature-level-fusion/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYV2VMKEIKMDXMKY2I7MDHLWZDCVDANCNFSM6AAAAAAVG7OGWA . You are receiving this because you are subscribed to this thread.Message ID: <zidu07/multimodal-biometric-verification-of-face-and-voice-using-feature-level-fusion/issues/1 @github.com>

a1once commented 1 year ago

@zidu07 thank you for your response! I have no idea what to search for. I need triplet_loss_trained_model.pth file, is this a publicly available trained model? Is it necessary that it was trained on audio?

Or could you please please search for this file in your drive or locally? It's super needed for me

zidu07 commented 1 year ago

Hi

So I couldn’t find the file, I might have accidentally deleted it cuz I can only find the shortcut in the drive the original file was deleted. So this file is model that using Siamese network and follows the one shot learning protocol. You may read up on this.

The closes file I could find on GitHub is https://github.com/sainimohit23/FaceNet-Real-Time-face-recognition/blob/master/train_triplet.py

On Fri, 24 Feb 2023 at 8:40 pm, a1once @.***> wrote:

@zidu07 https://github.com/zidu07 thank you for your response! I have no idea what to search for. I need triplet_loss_trained_model.pth file, is this a publicly available trained model? Is it necessary that it was trained on audio?

Or could you please please search for this file in your drive or locally? It's super needed for me

— Reply to this email directly, view it on GitHub https://github.com/zidu07/multimodal-biometric-verification-of-face-and-voice-using-feature-level-fusion/issues/1#issuecomment-1443823415, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYV2VMIZL2MFX3XS6Y753X3WZDFN5ANCNFSM6AAAAAAVG7OGWA . You are receiving this because you were mentioned.Message ID: <zidu07/multimodal-biometric-verification-of-face-and-voice-using-feature-level-fusion/issues/1/1443823415 @github.com>

zidu07 commented 1 year ago

So if you were to go under the cosine distance function where we give the path of the file you can see that it’s used to extract embedding from an audio file

On Fri, 24 Feb 2023 at 9:10 pm, test1 code @.***> wrote:

Hi

So I couldn’t find the file, I might have accidentally deleted it cuz I can only find the shortcut in the drive the original file was deleted. So this file is model that using Siamese network and follows the one shot learning protocol. You may read up on this.

The closes file I could find on GitHub is

https://github.com/sainimohit23/FaceNet-Real-Time-face-recognition/blob/master/train_triplet.py

On Fri, 24 Feb 2023 at 8:40 pm, a1once @.***> wrote:

@zidu07 https://github.com/zidu07 thank you for your response! I have no idea what to search for. I need triplet_loss_trained_model.pth file, is this a publicly available trained model? Is it necessary that it was trained on audio?

Or could you please please search for this file in your drive or locally? It's super needed for me

— Reply to this email directly, view it on GitHub https://github.com/zidu07/multimodal-biometric-verification-of-face-and-voice-using-feature-level-fusion/issues/1#issuecomment-1443823415, or unsubscribe https://github.com/notifications/unsubscribe-auth/AYV2VMIZL2MFX3XS6Y753X3WZDFN5ANCNFSM6AAAAAAVG7OGWA . You are receiving this because you were mentioned.Message ID: <zidu07/multimodal-biometric-verification-of-face-and-voice-using-feature-level-fusion/issues/1/1443823415 @github.com>