Quantcast
Channel: The Echo Nest Developer Forum
Viewing all 1582 articles
Browse latest View live

song/identify bucket=id:whosampled & bucket=tracks no result


song/identify bucket=id:whosampled & bucket=tracks no result

$
0
0

It's just that we don't have "I'm In It" in our rosetta mapping. A query for "Touch The Sky" works fine and returns the right ID. Depending on the partner our Rosetta mapping may be slightly out of date from their database.

Also, please have some patience, the Echo Nest does not normally work at 4am on a Sunday.

http://developer.echonest.com/api/v4/song/identify?api_key=XXX&format=json&results=10&code=eJwdVIsVAyEIW-kA5TOOou4_QpO-9nEYMUJEP43v-zQWTF6YGjCL2Ka3G6bp3YR5MPbR6EeDWRv0hsNMeo4ZC5pkXDKkOEtSW8TIbOdvOHsRTJJvCD2FP4xDw-wYHM6kQX7DQTOY7kga8o3mskNzGfyw2_wwO4We0ZB_DhDMyQnmN5OG9c7ixGJcM-7QI9W8GDt--NNTrHD7Gw4HPZL6ZIgnDSeCHvNzaupk9sWJTW8TOx8NhxTWH6iCGzGzL5hzMOcYWBbMOZwYp8n-RdEruEH62MQobFCNaK6gxHGwW1wOuUdS6RQsS0qc3CMnJ1h6cs_kuSWZc3G4sDaZeLIZksys6ssHU8y5ZNBgopRDtkSRtJh48dyKzMVzIzEM44oexjIfjkBaBCdzPGzrGADtjtsln454BRAKlbw7bE1pgFh4Cx-zwsI1gffFQrTprQB4bN9_8LZx0bv9QvsYIEPr7e8K1vfTTEusB-VSgBpRciHD7jp3hLXoA6jy3jc2t854SHc-xMmTjBmIE0-XMxEqcwHUqurpLHC5h4Dvm46yD_gAssQBPtt7Rh-xse89BPfRt7e9ie_g953A1wvfN3Ffz8LXcPL1vDC8lRoWqx7aAff33BUHda8EqBgoW-DkXtWbejgAm0Y9UsY9g3qcPQ3_L8T8jIsQ6XG2qIEoRLxGXhC9nlDLFLEPoKqsM16reaZvXLvKs1RsOgo3l61nlpet9t7T7dy7AUr7i71aRUL2ucgnx2zXfthdRMVZu6G9cq5-AsYPoClkQ3OLj07IhiObBVDnspw-sH_h5ok1959L6z3B_ngTIFu7PNcEKLbj4JpiAyjwHFX3BKjoYFQ97O01zl-qd2-Z9dprv82G1QUdcBcmr5qaaaLv0OTZvXTuOgC_q-U3BlKzXBO83tlXdRtHBubVF4-jKdLYBp2_3g4ZdWSbyh0nC3VVR_FhRZMHQKvNzm-ze5FxQMd-G6A3dLQrX08cMljWMIDQrbS-RlUodCBpVIVCpZE0QOiFd2VSN5xrJ95lfHGuOb_HLw4Jahw893VvTsFlRA9ujcTz1whPXN8yQXM6wE8s0LGJU8W7fEJQfA_st4ZrhbDyAog-zD4Lz76gP81-CSqfDQ==%22&bucket=tracks&bucket=id:whosampled

Bug in Artist documentation

$
0
0

Paul, you're wrong.

Artist genre

Calling...

http://developer.echonest.com/api/v4/artist/genres?api_key=API_KEY&name=weezer

Returns...

{"response": {"status": {"version": "4.2", "code": 0, "message": "Success"}, "terms": {"genre": "rock", "sub_genre": "alternative rock"}, "id": "AR633SY1187B9AC3B9", "name": "Weezer"}}

Bug in Artist documentation

$
0
0

My dropbox screenshot doesn't seem to want to embed, so here's the link to reference what I was talking about

https://www.dropbox.com/s/u7ud6mjdusj6sjz/echonest-artist-genre.png

Bug in Artist documentation

$
0
0

musicnews - no, I'm not wrong. Indeed, artist/genres is returning some data, but this is a non-documented and non-supported call and the data it returns should not be counted on to be correct, nor should you expect that method to always return data. To get the genre for an artist use the genre bucket like so:

http://developer.echonest.com/api/v4/artist/profile?api_key=YOUR_API_KEY&id=ARH6W4X1187B99274F&bucket=genre

The documentation in your screenshot is describing the artist bucket:

Bucket documentation

Unable to upload songs from computer

$
0
0

Hi there,

I'm using EchoNest to do a timbre analysis. When I use Code 1 (see below), I'm not able to make it identify the URL I type on my .txt file. The URL I have copied comes from dropbox and/ or soundcloud. Along the same lines, when I use Code 2 to upload the song from my own computer (.wav), I get the following: "Exception: Codegen binary not found." I have no idea what that means. Does anyone know how to resolve this issue? I'm not a computer science guy and dealing with this has been painful and time consuming! Is there anyone out there who can help me out?

thanks in advance,

CODE 1 lines 65-80

# For each line in the input file, extract timbre data from that url
for line in inputFile:
line = line[:-1]
(url, outputFile) = line.split("")

"""
This should find the specified file. Make sure it's in the same folder, or you'll have
to specify the specific folder
"""
t = track.track_from_url(url)

song_search = song.search(title=t)
song_title = song_search[0]

url = song_title.audio_summary['analysis_url']
extractTimbre(url, outputFile+".csv")

Code 2 lines 65-81

# For each line in the input file, extract timbre data from that url
for line in inputFile:
line = line[:-1]
(url, outputFile) = line.split("")

"""
This should find the specified file. Make sure it's in the same folder, or you'll have
to specify the specific folder
"""
#t = track.track_from_filename("url")

#song_search = song.search(title=t)
song_search = song.identify(url)
song_title = song_search[0]

url = song_title.audio_summary['analysis_url']
extractTimbre(url, outputFile+".csv")

Bug in Artist documentation

$
0
0

Thanks for the clarification. Why is /artist/genres returning data then? Will this be phased out?

Bug in Artist documentation

$
0
0

One other question, why is this genre which should obviously be R&B returning "R Neg B" ?

#


Bug in Artist documentation

$
0
0

R-Neg-B is a particular different thing. http://en.wikipedia.org/wiki/PBR%26B

Bug in Artist documentation

$
0
0

Learn something new every day. Thanks.

Upgrade rate limit fails with internal server error

$
0
0

Hei,

I have been trying to upgrade the rate limit of my account, but it fails with:

500: Internal Server Error

Are you having problems with your service?

In the partially pre-filled form i add country and region and the description.


Øystein

Unable to upload songs from computer

$
0
0

jkmm84,

If you have audio in a URL that you want analysis for, you do not need to do song.search or song.identify. Make a Pyechonest Track object from the URL using:

from pyechonest import track
t = track.track_from_url(url)
t.get_analysis()

Now your Track object has the full audio analysis, and you can get all the segments along with timbral information by iterating over t.segments. Uploading a track in this way will also do an implicit song/identify and fill in the SO ID.

The reason you are getting a Codegen error on the second example is that song.identify expects an ENMFP codestring (fingerprint). If you pass a URL or filename, it expects you to have the Codegen binary available to create the fingerprint. If this doesn't make sense to you, just ignore song.identify and stick with track_from_url() or track_from_filename().

~David~

Getting error_code 5 for some tracks.

$
0
0

I am using echonest API to create dynamic playlist. When i fire following request from device it gives code = 5. But if i fire same request from browser it give message = "success". Can anyone please tell why response different on device than browser.

Request-

http://developer.echonest.com/api/v4/playlist/dynamic/create?api_key=TAFYO7QHYKWQE791F&artist=Nakash%20Aziz,Ritu%20Pathak&song_id=spotify-WW:track:1At6Fbe8RxXE6tKKJtMbyW&bucket=id:spotify-WW&limit=true&type=artist-radio&adventurousness=1

Response on device

{ status = { code = 5; message = "The Identifier specified does not exist"; version = "4.2"; }; };

Getting error_code 5 for some tracks.

$
0
0

Hi Rahul - there shouldn't be any differences whether the request comes from a browser or a device. Double check to make sure that the requests are identical. Perhaps the trouble is with the URL encoding of the artist name? Also note that for artist-radio you do not need to pass in a song_id and it will not be used so you can omit the song_id parameter. -- Paul

major / minor key mode=0 or 1

$
0
0

I have tested a couple of songs. And it seems to be consistent that Mode 0=minor and 1=major, but according to http://developer.echonest.com/raw_tutorials/faqs/faq_04.html it is supposed to be the opposite.


major / minor key mode=0 or 1

$
0
0

hviderusland,

You are right. Mode 0 is minor and 1 is major. That FAQ page needs a refresh. Thanks for bringing to our attention.

David

Playlist hotttnesss sort not working

Can I locally compare echoprint audio file fingerprints?

$
0
0

I am also interested in this. especially if it would be a library which compares two audio files. is there anything like that in you project? that would help me a lot.

Getting error_code 5 for some tracks.

$
0
0

Hi Paul, Thanks for the reply. Yes it worked. The problem was in URL encoding in artist name.

I got another weird problem. Here is URL -

http://developer.echonest.com/api/v4/playlist/dynamic/create?api_key=TAFYO7QHYKWQE791F&artist=KK,Pritam,Shefali%20Alvares,Yo%20Yo%20Honey%20Singh&song_id=spotify-WW:track:0PSALeWN6XzdEtNEDl8BD5&bucket=id:spotify-WW&limit=true&type=artist-radio

If i shuffle artist names, it works properly. but the sequence mentioned above returns - { status = { code = 5; message = "The Identifier specified does not exist"; version = "4.2"; }; }

and yeah now the response is same on both device and browser. Can you please help me out. How can sequence of Artist_Name affects response. If so then what would be the ideal sequence. I am damn sure i m missing something. Thanks in advance. :)

Getting error_code 5 for some tracks.

$
0
0

Rahul - you need to give each artist with a separate artist param like so:

http://developer.echonest.com/api/v4/playlist/dynamic/create?api_key=YOUR_API_KEY&artist=KK&artist=Pritam&artist=Shefali%20Alvares&artist=Yo%20Yo%20Honey%20Singh&bucket=id:spotify-WW&limit=true&type=artist-radio

Also, again, you won't need the song_id param.

P

Viewing all 1582 articles
Browse latest View live