References and Documentation

[chrisport]

This repo looks really interesting, it's actually the only one I found about Audio Fingerprinting in Golang. I'm currently working on a small private project to index and recognize music.
Would you mind to share some references about the algorithms and approach that you are following?
Is it just a personal playground or are you planning to continue to work and maintain it in the future?

[snuffpuppet]

Hi Christoph, Thanks for the email. This is just a passion project at the moment. Part of a project I'm (slowly :) doing with my friend to display subtitles to movies for the hard of hearing. At this stage I have been following the advice of many articles written about Shazzam but those algorithms seem to have some problems. The fingerprint hit rate is quite low and they don't account for limited microphone and speaker sensitivities. There are many articles written Shazzam out there following the paper written by one of the founders. If you want to read something cool, check out chromaprint: https://oxygene.sk/2011/01/how-does-chromaprint-work/ The spectre code is quite configurable which probably makes it a little difficult to follow. It is the result of a lot of experimentation so heaps of different ways of doing something, most of which are now deprecatated but still supported by the code. Anyway, feel free to fire me any questions you may have. I've started working on it again so hope to have a better version working soon Adam

…

On 19 Aug. 2017 18:57, "Christoph Portmann" ***@***.***> wrote: This repo looks really interesting, it's actually the only one I found about Audio Fingerprinting in Golang. I'm currently working on a small private project to index and recognize music. Would you mind to share some references about the algorithms and approach that you are following? Is it just a personal playground or are you planning to continue to work and maintain it in the future? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFELbducNsDIoaxD5dFwv4TTXxEkTk12ks5sZqOCgaJpZM4O8RjB> .