search

facebookresearch / StarSpace

Learning embeddings for classification, retrieval and ranking.
MIT License
3.95k stars 531 forks source link

Does this algorithm work on audio data?Can any body give me an example to use this algorithm for audio data classification? #213

Closed SibtainRazaJamali closed 5 years ago

LAV42 commented 5 years ago

If you obtain a feature vector for your clip through a neural network, you could use these features and do just like in the image example: https://github.com/facebookresearch/StarSpace#imagespace-learning-image-and-label-embeddings

ledw commented 5 years ago

@SibtainRazaJamali yes, as @LAV42 mentioned, you can use it similar to the image example. Let us know if you have any other questions.

SibtainRazaJamali commented 5 years ago

Yes @LAV42 and @ledw Thanks for kind response , i can implement this but can you suggest me ,if there is any specific algorithm to extract important features from audio signal. Since spectogram of audio signal contains different features as compared to normal image.

ledw commented 5 years ago

@SibtainRazaJamali Apologize for the delay in replying. I am not familiar with algorithms to extract features from audio signal.