Music streaming, playlist creation and streaming architecture
A system and method for making categorized music tracks available to end user applications. The tracks may be categorized based on computer-derived rhythm, texture and pitch (RTP) scores for tracks derived from high-level acoustic attributes, which is based on low level data extracted from the tracks. RTP scores are stored in a universal database common to all of the music publishers so that the same track, once RTP scored, does not need to be re-RTP scored by other music publishers. End user applications access an API server to import collections of tracks published by publishers, to create playlists and initiate music streaming. Each end user application is sponsored by a single music publisher so that only tracks capable of being streamed by the music publisher are available to the sponsored end user application.
1. A method for selecting a song, comprising:
selecting the song based on a computer-derived comparison between a representation of the song to known similarities in representations of other songs, wherein the known similarities in representations of other songs is based at least in part on a human-trained machine using the representations of the other songs, wherein the representation of the song is based on isolating and identifying frequency characteristics of the song, wherein the representations of a plurality of the other songs are based on a human listening to each of the plurality of the other songs in order to isolate and identify frequency characteristics of the plurality of the other songs, wherein the frequency characteristics of the song correspond to one or more moods of the song and the frequency characteristics of the plurality of the other songs correspond to one or more moods of the plurality of the other songs, and wherein the selection is based on the similarity between the one or more moods of the song and the one or more moods of the plurality of the other songs.
2. The method of claim 1 , wherein the representation is an audio fingerprint.
3. The method of claim 2 , wherein the audio fingerprint is based on a spectrogram of the song.
4. The method of claim 3 , wherein the audio fingerprint includes a plurality of sub-images.
5. The method of claim 4 , wherein the plurality of sub-images represent frequency of the song over time.
6. The method of claim 1 , further comprising sampling the representation.
7. The method of claim 1 , wherein the representation is based on frequency data, further comprising analyzing the frequency data to capture intensity differences represented by the frequency data.
8. The method of claim 7 , further comprising filtering the frequency data to create a digitized representation based on the frequency data.
9. The method of claim 1 , wherein the representation is a static visual representation of the song.
10. The method of claim 9 wherein the static visual representation is a spectrogram.
11. The method of claim 1 , further comprising checking the one or more moods of the song against at least one other mood identification technique for a level of correlation between the one or more moods of the song and the one or more moods of the plurality of the other songs.
12. The method of claim 1 , wherein the frequency characteristics of the song and the frequency characteristics of the other songs are based on a melody, a rhythm and a harmony of the song and the plurality of the other songs.
13. The method of claim 1 , wherein the one or more moods represent multiple moods in the song and the plurality of the other songs.
14. The method of claim 13 , wherein the multiple moods of the song are represented by percentages of the one or more moods.
15. The system of claim 1 , further comprising checking the one or more attributes of the song and the one or more attributes of the other songs against at least one other mood identification technique for a level of correlation between the one or more attributes of the song and the one or more attributes of the other songs.
16. A system for selecting a song, comprising:
sample frequency data from the song;
generate a representation of the sampled frequency data;
filter the representation of the sampled frequency data to capture intensity differences represented in the representation to generate a digitized representation that corresponds to one or more attributes of the song;
compare the digitized representation to known similarities in digitized representations of a plurality of other songs having one or more attributes of the plurality of the other songs, wherein the known similarities in the digitized representations of the plurality of the other songs are based at least in part on a human-trained machine based on a human listening to each of the plurality of the other songs in order to isolate and identify frequency characteristics of the each of the plurality of the other songs, wherein the digitized representation is based on frequency characteristics of the song, wherein the digitized representations of the plurality of the other songs is based on the frequency characteristic of the each of the plurality of the other songs, wherein the frequency characteristics of the song and the frequency characteristics of the each of the plurality of the other songs correspond to the one or more similar moods of the song and the each of the plurality of the other songs; and
select the song based on the one or more similar moods.
17. The system of claim 16 , wherein the digitized representation and the digitized representations are static visual representations of the song and the each of the plurality the other songs.
18. The method of claim 16 , wherein the one or more attributes of the song and the one or more attributes of the other songs are based on a melody, a rhythm and a harmony of the song and the other songs.
19. The method of claim 16 , wherein the one or more moods of the song represent multiple moods in the song.
20. The method of claim 19 , wherein the multiple moods of the song are represented by percentages of the one or more moods.