worldveil / dejavu

Audio fingerprinting and recognition in Python
MIT License
6.35k stars 1.43k forks source link

How do I use it for 100w music #266

Open Novicei opened 2 years ago

Novicei commented 2 years ago

Thank you for reading. During my experiment, I found that only 5 pieces of music will generate 500,000 records in the database. What if I want to use 1 million pieces of music? I think Shazam's paper should be able to support large-scale music retrieval.

alffykasum commented 1 year ago

hello @Novicei this is only my view but Dejavu is a software project that allows you to recognize audio fingerprints in a large music collection. It is designed to be scalable and efficient, so it should be able to handle a large number of audio files. However, there may be limits to how many audio files it can process at once, depending on the hardware and software setup you are using.

One way to scale up Dejavu to handle a large number of audio files would be to distribute the work across multiple machines. This would allow you to parallelize the processing of the audio files, potentially increasing the speed at which they are processed. You could also consider using cloud computing resources, such as Amazon Web Services (AWS) or Google Cloud Platform (GCP), to scale up your processing power.

Another option might be to use a different music recognition software that is specifically designed to handle large scale music retrieval. Shazam, as you mentioned, is one example of a service that can handle a large number of audio files. There are likely other options as well, depending on your specific needs and requirements.