That song no longer appears to have tracks in 7digital-US. Track rights change all the time and our catalog refreshes to reflect that.
David
↧
Bug in a certain song?
↧
Analysis not found
GipsyWolf,
I see TRAQONU1451D8D7F8A as created for that MD5 (afd09da8b0aeda6557dce2a3e1cb20ba), so I'm not sure why the analysis was not created. Is this still happening with the latest version of Pyechonest?
David
↧
↧
Bug in a certain song?
Ah, OK. Just a timing coincidence. Thanks.
↧
TasteProfile problem updating (adding) songs
Ok, so it turns out this was all a problem with caching iOS does with url requests. I created a pull request for enios on github that fixes the issue.
↧
Is there a way to get a list of EN song dataset
Hi plamere,
Thanks for the reply. would there be some sample dumps available for seeding ?
des
↧
↧
No song data
Also: I've edited your question to remove your API key, so it doesn't get stolen and abused.
↧
No song data
The is I am using is the SO.... is as returned by the first request.
Did you mean the other way around? The song search only returns a SO... type number I think..
↧
No song data
Right, that's my point. Since the entity in question is a song (not a track), you need to call song/profile rather than track/profile:
http://developer.echonest.com/api/v4/song/profile?api_key=XXX&id=SOJOAOT1458B54F8BC&bucket=audio_summary
-Eli
↧
No song data
I see it now, sorry - I need to change the word "track" to "song" in the second request. I have the ID the right way - not the request.. Thanks!
↧
↧
Artist confusion on playlist generation
Echonest is confusing the band "timeflies" with the rock band "time flies" as seen in the request below. The requested artist id points to timeflies, but many of the results are either from the rock genre, or from the band "time flies".
http://developer.echonest.com/api/v4/playlist/static?api_key=API_KEY_HERE&artist_id=spotify:artist:6dC0rIJNLSFZwqckLgXJ8p&format=json&results=100&bucket=id:spotify-WW&bucket=tracks&limit=true&type=artist-radio
↧
Server Error Uploading MP3
Tried to upload two different songs, both times received an error page saying "Server Error - There was an error processing your request."
EchoNest has worked for us before, but is not today. We just recreated our developer account and are using the new API key, so it can't be that. Tried a different song that I knew worked before and that still worked. The only difference I can identify between the songs is that the one that worked was encoded with iTunes 6.0.4.2 and the two that didn't are encoded with iTunes 7.6.1.
Help?
tri nam | nha dam | tri nam tan nhang | nam da | trị nám tàn nhang | đồi mồi | kem chống nắng
↧
Profile read with Spotify ids
Hi,
I'm creating an taste profile and updates it with track_id = spotify-WW:track:1XSNeHpiFie7bv5vL2E2ee (for example).
When I later wants to make a taste profile read to get the audio summary things get messy.
For some songs I get the audio summary for a different track, like in this case San Cisco - Beach and not Boy and Bear Beach.
{
"artist_id" = ARYUOHX131529ED7A2;
"artist_name" = "San Cisco";
"audio_summary" = {
acousticness = "0.639028";
"analysis_url" = "http://echonest-analysis.s3.amazonaws.com/TR/Wr5HbQVTse-oWPq3hRc5PQQYokzgVmcunid7LsUMETL8rHtygaZXPuZwAYLYc8rnsYiw_5o1yBEX4sYHs%3D/3/full.json?AWSAccessKeyId=AKIAJRDFEY23UEVW42BQ&Expires=1403526879&Signature=UYKGdcVKI2tvQUX3Uig6YXCYIzc%3D";
"audio_md5" = 8bc3a34cec720dfc80fea8c27935b6c8;
danceability = "0.441795";
duration = "220.96063";
energy = "0.627918";
key = 9;
liveness = "0.114956";
loudness = "-7.19";
mode = 1;
speechiness = "0.036524";
tempo = "184.962";
"time_signature" = 4;
valence = "0.264683";
};
"date_added" = "2014-06-04T13:20:18";
favorite = 1;
"foreign_id" = "CAHMGGP1464CAF6B18:song:spotify-WW:track:1XSNeHpiFie7bv5vL2E2ee";
id = SOGUVHV13B1F7871ED;
"last_modified" = "2014-06-04T13:20:18";
request = {
"item_id" = "spotify-WW:track:1XSNeHpiFie7bv5vL2E2ee";
"song_name" = Beach;
};
"song_id" = SOGUVHV13B1F7871ED;
"song_name" = Beach;
}
Whats seems to be wrong? Is it a bug?
↧
Profile read with Spotify ids
Runbeat - the spotify-WW ID space has been deprecated. You should use the new, spiffy and higher quality 'spotify' id space. As you can see with this query:
http://developer.echonest.com/api/v4/song/profile?api_key=YOUR_API_KEY&format=json&track_id=spotify:track:1XSNeHpiFie7bv5vL2E2ee&bucket=audio_summary
the mapping in the new ID space is correct.
Paul
↧
↧
Profile read with Spotify ids
Wooho, thanks! Works
↧
tatum / segment relationship
To render say, just the last hit of a track,
audiofile = audio.LocalAudioFile(input_filename)
audiofile.analysis.tatums[-1]
returns the final tatum, but trailing tail is cut off.
audiofile.analysis.segments[-1]
plays the complete final hit, including fade and silence.
The problem is that the final segments and tatums start and end times do not line up:
segment (303.36 - 303.50) segment (303.50 - 308.96) #END is 4 seconds later than tatum end
tatum (303.41 - 303.69) tatum (303.69 - 303.96)
In order to append the final segment to a track assembled of bars and tatums, is it possible to override the end time of the final tatum or shift the start time of a segment to match a tatum?
↧
tatum / segment relationship
Tatums and segments are independent features that only relate to each other on a perceptual level. It is normal they don't align exactly. As a result, tatums might stop before the end of the track, while there's always a final segment up to the very end. It's up to you to manage the logic of your code to handle the relationship between the final segments and the last tatum: there's no built-in direct bridge between the 2 layers. You can always override the time values manually if that works for you, but that will affect what the renderer does. Hope this helps.
↧
tatum / segment relationship
Thanks Tristan.
Pardon my PyInexperience, but when I tried to reassign audiofile.analysis.segments[-1].end, it returned an AttributeError: can't set attribute.
Here's the code tat displays my ignorance, if anyone wants to enlighten me.
track.tatums[-1].end = track.segments[-1].end
Peace.
↧
↧
how to generate codegen.exe?
Hi friends i have just started compiling codegen window using the readme given in the file. following the youtube link http://www.youtube.com/watch?v=NmYfPea4ELU i have managed to generate the 2 DLL files.
the readme says that i need to load the codegen.sln and i try to compile it but i have the following error message
" error C1083: Cannot open include file: 'tablib_config.h'. No such file or directory "
also as mentioned in the readme i am supposed to see a codegen.exe but i cannot find it anywhere in the folder. Is it automatically generated after i build codegen.sln?
thanks for reading and thanks for your help
↧
Updating Taste Profile Problem
Hey guys, just having another problem updating the taste profile, specifically in regards to the Rdio buckets. The read profile works. However any song added to the taste profile after ID of 30,000 does not give any returns. Please can someone take a look. Here is an example. As you can see any songs with an ID of around 3100 display the Rdio, however for the songs around the 31000 value, no bucket information is displayed? This continues for the entirety of the results. Can anyone help?
http://developer.echonest.com/api/v4/tasteprofile/read?api_key=YQOCUCOLLPTOECWLK&format=xml&id=CAKKFGP146A91ACA17&results=50&start=22050&bucket=id:rdio-AU&bucket=tracks
We have tried everything. We even tried creating a new taste profile and added the extra songs with ID's of more than 30000, however they still give no Rdio bucket outputs.
↧
Updating Taste Profile Problem
Does anyone have any ideas? It makes no sense. I've figured out how to make it simpler to understand. If you look at the following URL, you can see where the tracks stop outputting. At a track 'start' of 21908, the song "I Gotta Feeling" (and every song before it) works fine, and outputs track information with Rdio buckets. However every song after it (with a song ID greater than 30963, no track information is displayed. Which means we can't query any buckets.
http://developer.echonest.com/api/v4/tasteprofile/read?api_key=_API_KEY_&format=xml&id=CAKKFGP146A91ACA17&results=20&start=21908&bucket=id:rdio-AU&bucket=tracks
Thanks guys.
↧