Is anyone doing static playlists that are generated from their own music collection?
↧
Seeded static playlists
↧
Seeded static playlists
I'm not using it in a product, but I do have a dev tool which does return static playlists.
You should check your profile's status. Make sure its analysis has finished. How did you populate it? I've seen problems trying to add data with non-latin characters to the profile. Might just be a matching issue, but still.
↧
↧
jEN API: get song/track with Rdio ID
You want to use song/profile, with a parameter that looks like "track_id=rdio-US:track:the id you got from Rdio". Then pass in buckets for the various Echo Nest data you want.
↧
Seeded static playlists
There is 9411 items in the profile, but it is just a general profile. I'm going to try recreating it as a song profile and see how this works.
↧
Can I fetch albums?
Thank you for your comments.
Do you have a plan to expose album info?
↧
↧
Fresh API key now working
Hi Francois - we are taking a look -- Paul
↧
Similarity metrics
Hi,
Similarity of work is certainly a very complex issue, nonetheless the results given by similar can be extremely weird. One way to solve it would be, as it is solved in terms, to weight the similarity. Possibly, this weight could be not necessarily a single number but a vector of x in [0,1] or arbitrary real numbers depending on the metrics used, and making explicit the different notions of relatedness.
Could such a weighted_similar be implemented? The issue was already raised in thread http://developer.echonest.com/forums/thread/49
↧
Can I fetch albums?
yes, we have a plan, but I can't make any official announcements about it yet. -- Paul
↧
Time Offset of detection in the audio stream?
+1! I'm analyzing a radio streaming, so I take 20 seconds and get it analysed by calling /track/upload.
If that time offset is given, I just need to call API again after current song ends.
↧
↧
Estimate for /track/upload
Would be nice if after a /track/upload you could add a variable indicating when to come back and check for /track/profile. Could be:
- A simple info like 'How long the last processed song waited on line'
- Maybe a more complex method could count how many songs are queued and multiply it by an average processing time, maybe taking in account each fragment duration.
That could help to avoid call your API as my request is unlikely processed.
↧
Estimate for /track/upload
thanks for the feedback. Good suggestions. -- Paul
↧
adding a list of songs to a taste profile?
Thanks for the help. I figured out my mistake: I was failing to submit an item_id with each update. I was just sending an artist_name/song_name pair, and understandably needed a unique ID to keep everything straight.
I'm now building away with Taste Profiles. This looks like a very promising part of the service I hadn't explored before!
↧
Can I fetch albums?
Thank you very much.
↧
↧
Can't get multiple Buckets to work with $.getJSON
Hi, I am running into some trouble getting the buckets to work. Every time I try to have multiple buckets and pass them in as an array like examples I've seen, I get an ''Access-Control-Allow-Origin' header is present on the requested resource.' Error.
I looked at the request URL and it seems that if I try to use the array it encodes the brackets, and I cannot get any request with that encoding to go through. (Ex: http://developer.echonest.com/api/v4/artist/profile?api_key=hidden&bucket=news%26bucket%3Dartist_location&format=json&id=ARE602H1187FB3B8F8
Here's a basic example of what I'm trying to do:
var url = 'http://developer.echonest.com/api/v4/artist/profile'
var args = {
apikey: apiKey,
//bucket: 'news', --This works
bucket: ['news', 'blogs'], --This does not
format: 'json',
id: artistsEchoNestId
};
$.getJSON(url, args, function (data) {
console.log(data);
});
Any suggestions would be appreciated, thanks!
↧
Can't get multiple Buckets to work with $.getJSON
Chris:
add this to your init code:
jQuery.ajaxSettings.traditional = true;
Paul
↧
Can I batch fingerprinting?
Hi everyone!
We want to program a music player that allows users to clean up the tags of their music library. To do so, we would like to use the amazing fingerprinting capacity of ENMFP .
As we looked at the echonest API, it appeared to us that identifying tracks can only be performed through individual requests.
PROBLEM : a user can have in his music library thousands of tracks (and more!). Identifying his whole music library by individual requests, would result in serious HTTP flooding. If we consider now that our music player could be used by many users at the same time, the htttp flooding would be MASSIVE.
Any idea of a reasonable way to perform massive fingerprint identifications? Would it be reasonable to set up cache servers to store the fingerprints of our users and thus limit the identification queries to echonest? If we do so, wouldn't be echoprint a more appropriate solution than ENMFP?
Thanks in advance for your answers! Paul
↧
Can I batch fingerprinting?
Yes, you can. Our docs are not great about this but you can batch multiple FP queries in a single POST call to song/identify. The query parameter in POST mode can take a list of codes, the same one that the ENMFP and Echoprint codegens output given multiple file input. Here's an example:
caro:Geogaddi bwhitman$ cd ~/Music/iTunes/iTunes Media/Music/Boards Of Canada/Geogaddi/
caro:Geogaddi bwhitman$ ls -1 *.mp3 | ~/Downloads/ENMFP_codegen/codegen.Darwin -s 10 30 > codes.json
caro:Geogaddi bwhitman$ curl -F "api_key=XXXX" -F "query=@codes.json""http://developer.echonest.com/api/v4/song/identify"
↧
↧
Fingerprinting: ENMFP VS Echoprint + Echoprint contribution
Hi everyone!
We want to program a music player that allows users to clean up the tags of their music library. To do so, we would like to use the amazing fingerprinting capacity of ENMFP... Or echoprint if the latter is more appropriate for batch processing.
We have quite a huge dilemma regarding the choice of ENMFP vs EchoPrint: the first offers a wide base and some codegen issues, the latter is lightweight, offers OTA but with a still quite-reduced database. As we are a very small team with extremely tight resources, we can not really afford a double compatibility echoprint/ENMFP.
As ohrstrom, we do want to contribute to echoprint if we use it. Do you have any release date for the tool you are working on in order to contribute to echoprint? I guess the availability of such a tool would seriously impact the echoprint DB growth rate...
Do you have any idea of the (approx.) echoprint DB growth rate? Any advice about choosing between ENMFP vs echoprint in our case?
Thanks in advance, - Paul
↧
Can I batch fingerprinting?
thanks a lot for your answer!
↧
Youtube Channel
Greetings!
I've been searching through the forum and the API doc, and yet I did not mange to find how to retrieve a Youtube official channel using an artist id (eg. Radiohead official channel). Echonest seems to amazingly provide many information such as youtube videos for an artist, but yet not the official channel
Have you considered to integrate youtube channels in echonest?
↧