Does this support text independent speaker identification?

pannous / tensorflow-speech-recognition

🎙Speech recognition using the tensorflow deep learning framework, sequence-to-sequence neural networks

Other

2.17k stars 638 forks source link

Does this support text independent speaker identification? #57

Open channae opened 6 years ago

channae commented 6 years ago

Hi,

Does this support text independent speaker identification? Appreciate if you could share some insight.

Thanks,

Mayanksoni20 commented 6 years ago

I think it should . I am doing that right now , will share any information that i get. Essentially , try to train it on some words for all speakers and then check if it can identify other words

taewookim commented 6 years ago

Is there en audio embedding layer that can be used to compare (i.e. via eulicidian distance) to see if 2 voices are from same speaker?

channae commented 6 years ago

@Mayanksoni20 Awesome, Please do keep posted. Thanks

4emkay commented 6 years ago

Even i am facing the problem in improvising the accuracy for text independent speaker recognition. Please do post the updates Thank you.