That makes sense, and I do like having a number for more fine-grained search control. Looking closer, it looks like the decision was to use just one of the nodes for traits that can be mapped on a binary spectrum--e.g. the user can assume that if tonal > 0.5, it's probably atonal, and factor that into their searches--while those that can't, like "rhythm", just take the value node. Is that more or less the logic? That certainly makes sense for the tonal/atonal thing, now that I know how it works (documentation would indeed be helpful--do you need help with getting that written out? how would I go about contributing to that?). I could see it being useful to have an option, maybe in the config, to use the "value" node instead if you'd prefer not having a sliding scale--but I don't know what kind of demand there is for that, tbh.
On a more personal note: are there plans to include the mood_mirex or timbre data points in the plugin? Cuz boy oh boy was I excited to see that AB has that info. On Thursday, May 26, 2016 at 4:50:37 PM UTC-6, Adrian Sampson wrote: > > Interesting point. I think it was probably written this way so that we get > a summary of the "tonal-ness" determined by AB. That is, we wanted a single > number that reflected as much information as possible; the assumption is > that if you want a binary decision, you can just ask "tonal > 0.5". > > In any case, we should do a better job of documenting that -- and make > sure that this strategy is actually useful for recovering the right > tonal/atonal decision. > > On May 26, 2016, at 2:35 PM, Hush Piper <[email protected] <javascript:>> > wrote: > > I may be entirely misunderstanding the plugin here, so apologize in > advance for that possibility! > > When using the AcousticBrainz plugin and viewing the results (e.g. with '$ > beet info -l tori amos sorta fairytale'), the actual values I'm seeing for > things like tonal or mood_electronic don't seem to match the high-level > results listed in the AcousticBrainz API data page: > > http://acousticbrainz.org/data > > An example is the track A Sorta Fairytale by Tori Amos. The value for the > tag "tonal" came out to 0.486311525106, as opposed to "atonal", which is > the actual value: > > > https://acousticbrainz.org/api/v1/e32badcb-4bb1-43ed-a7fb-d947c17042b4/high-level > > I haven't actually read over the code, but it looks like it's grabbing the > xml node highlevel > tonal_atonal > all > tonal, instead of highlevel > > tonal_atonal > value. I'm guessing it's not supposed to do that; the value > of just the tonal node on its own doesn't actually mean anything in this > context (especially since it's actually marked as atonal). > > Note that some of the keys are coming out fine, rhythm and > voice_instrumental among them. It's just the mood_ keys and a handful of > others like tonal_atonal. > > I can attach my config if needed. Here's the full tags for that track > (apologies for the length and mess, I also have echonest and rymgenre > going): > > /home/redacted/music/Tori Amos/Scarlet's Walk/02 A Sorta Fairytale.mp3 >> acoustid_fingerprint: >> AQADtEmyNEmSJEomyI6O_IZH4oliNJlRNrsR-kiW8_iSKD4e7ROyMlGgpUd4LcelB0-4OPh_7ELzILyQ5Dv-FFySJJdxKuiH5kmFXEJyg0224uuK012CH0ey5BnCHy990MdMO6iYC9aOW9YRXlD-4ouPL6KO5PgTFX2GZqkJ9seZSIY_JH4yCVly3rCPSskU3EpxI4WuLUVm9cil5NDyCaF5XPzxRDeucKlwDtPRPIfyhMTBKXswJnooPInh40mg6wKeKMcTJcevBnuP5Eb645l6XAofPCfqDc1zPAtx-cLEaQ-DO0Yv3BSeSFmGPkdz_Pg-uP1wlB3RvBzyI4nKHU48Hg4VFv0TnBuaZ8fU6HiX48d54WFWVLSJhssjXD1C50g4eP1RKY2Dp_IS5LkI5UGmacWf4iee40yOnCGRXHjg4VkGzsTkB4344gryUMUXaFq24FJydM2eEKp45Eez5kOdKccf_IefRcd3HDe-EFWOZImYCtGn4_zwHV9mNKOUFVVpPE0u_DgtoV2Okw9GOiueHAkZJRfCK6GIL_jwjyGeJ0j9QPmHXBKaRy96BcnkLEOWKBlxfYZxgj-2I2YEeRQRSjk8xgr6owmTFMk2XYi_LsQ3eFyMJyd-XPB_XMYb3LiLPkaycL0QMQtnnDx-Be1zeD8q6niW5GCZIuciJOQSJRbOHceDKuk4VOuRXHyQ6wuO5sxQPcW54fiPeMkPveGwG-kTnBPeI-KmDlSSa8zwD7o-YQqTHX2OBm0VQVdK5Iq54PzQlSeaqA_uBc8WFY59_MepRXiUHb9wJOwzxB1W5UTJ7HDMFBV_nMcz6TjFo4nyosofXMk0-DoShPlRhlGOv8FbKaAS5kNME8lhMkcpKiIaM0euBEmNNXkGzwrxNIp0XGYw5UfP8EiOmPngZC-x55iJMksSuJJyiE0uoU8LXHkSXOpQJ8nBREf844EuEk9SSfiD5gbzQE-u4ZuN_BSmnONQPcGVyXjzJJio448OcxnYov8xvsgjDarC5ciPXpnRVDxu48xa_IgVBUmz5_gGH1c-fMcVKYWfHnUT4Tye8ejDJGi4I3wiDbofhOoPJxqD24FnbeiTCH-OZ3gehImYS6gyfdAP_RSmPArC48dzhM-HywlUMVjlJIcmnfiDUklepNWH1mmOvHxw8tDxKFqFsBXh6EE_PMfzEE1FC5V0XLhKcWiKHt_xXCj1orGO_7gQkhP8REavHvpH_FqGpvoQ6szxaHhtnMqIPoEv9MyLR2XQxUfDebiUZ0THKQqaOIqLHz3C_Pig_WgmhxH0rah0OHnQnMbxQ2s3BmmPe4eXZRzgN0N_vMGbDxczPPrR5YdfVM-FCzfO4T5qHfksiDzSqMc33Id_-HLQn_h7PJGEPIvwQ3vR5Esp6CkeJQ_xLDgqMwjzRAau5EGe4IeWvUOZKMST486Cp2jGUDzKjPiDMzrCe4dHGg8f-CzeGM0bLrgffMnx8Kj0wM_QK8Z_3NbxbEdZHTfcvSh9ISeSEc3xC681lKlw9C8eNOOCNmZTlE_QGEXjZNGQS4euI0x2opsS4e4xM-hBhcqOLsZz7BdCLQ_E4ZaETzLk6ohvTFJ-_OCOnRcaEzVF4ce1C9eFpMkRSzoa3pFQOklx_DiPKs3hRUcl7cOYHx-Pfk-Qt0juHc2YiHiOm0xyPNNTIl0SSC3y41mEpOQRjlHgKEF5XDi4NBcqMw8i5YeuH_lyNFmOLgxuHr1yhGnyQVWiyAt-ItzhoI6PH2csNJMyC_2CG1eEyjkhqkdxnfgSdFR0_OIG10cpJXh25HkOLW4qnMrwC2kOPy9KSsyQZwtU6iLyZU_gdcKcB2-O-ziPnA-SMZFQ8mDQXCO4L3gw6UFDZQ9OGw0ZJDtT5HCT48oeXMGP9Bl0JshxkbiyVEG5J2iWMMCl6EiurCvS6Hiu4Q-6bvCTscNFODl-wVc4XMGv4D_SIJmPhz9-hrh1rBMbWBlDsImM2Imh54GuK3h65D-mDs13nDnuYP2RSPkRnVfwJGGKP2hO1DrqJIHPVKiUjUbZB45lJHOaI9-OF81PoY-Cq8Sz42PxC1eD8EiW_bjx48k19FnRXLpwpkKVZIfTxIqFOxvhK0JptkjWI8wZY6-OczryKCFzSMo33EfwD1UWfsgJPUh5o_qxltFwB_-weUfeQ7uSI-TxnMKTXQn-o88RplSOpJKPJ7mM2-gvNBXxRPnx6gluCy_aMXB0aCteXDuekBH6THjeIRmPPCMYn2jyLdjV4Ut2fMJzpBmnQ99S5CQjojkqJ8eTxURNoXmWsaj1FF6EKpEM1ZQUPPCPKyf-ox3RjAmPN0MvS0LJRJnwC-az4rKKUwiT9kimhFl0HA2ziSipNXiLxpGCWnmDOw-qhDqeJ_gxpQof_MdREX7G4yQjbIkWoTKlI0_QX9Cj-MRT4wrRrJeQO1tCvBPxzSPxp3DW7TgzET_uPGguoz-e6MOVqGieo3-GD08oC00aHqUUET-S0cLDWmhOB7pE3PmgJ22E5qFyVBtV5JGCt3lwZQx-Bb0On8GjMoHZD5-M5yEsNooQyhkuozm057iEnieaSVEyGvqOku0Qxy1-qeh5_AjDZxSSi4edGmU440dzJkIv4ceP5ziVoXxgqcJfol9xHF_Rl5IR5tB1VEQq8cfFBhcV9Bc-_HgmBfmSEc1R3J1EnMkT_LHwDO-JMDmLHniSB7mge2iWLUepaDku4mmIyoe4LPRR8tCPc0d47_Ay4szhh7DLoW8j3Dke5biyozyaL3lxnC-eRhdaj_CK88d_HMmRDr1wHb_xOcdz4jp-XEyFLusLZ1WO_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 >> acoustid_id: 20ac4c0d-b06f-42d1-8521-141591ae9f18 >> added: 2015-12-14 17:25:45 >> album: Scarlet's Walk >> album_id: 2087 >> albumartist: Tori Amos >> albumartist_credit: Tori Amos >> albumartist_sort: Amos, Tori >> albumdisambig: >> albumstatus: Official >> albumtotal: 18 >> albumtype: album >> artist: Tori Amos >> artist_credit: Tori Amos >> artist_sort: Amos, Tori >> artpath: /home/redacted/music/Tori Amos/Scarlet's >> Walk/cover.jpg >> asin: B00006L3YI >> average_loudness: 0.920467495918 >> bitdepth: 0 >> bitrate: 320kbps >> bpm: 153 >> catalognum: EPC 508782 2 >> channels: 2 >> chords_changes_rate: 0.065244846046 >> chords_key: G# >> chords_number_rate: 0.00126280344557 >> chords_scale: major >> comments: >> comp: False >> composer: >> country: DE >> danceable: 0.436882615089 >> data_source: MusicBrainz >> day: 28 >> disc: 01 >> disctitle: >> disctotal: 01 >> encoder: iTunes v7.6.0.29 >> ffmpeg: CRC=0x88d082be >> >> filesize: 13267551 >> format: MP3 >> gender: female >> genre: Rock >> genre_rosamerica: rhy >> grouping: >> id: 24397 >> initial_key: G >> key_strength: 0.767497420311 >> label: Epic >> language: eng >> length: 5:29 >> lyrics: >> mb_albumartistid: c0b2500e-0cef-4130-869d-732b23ed9df5 >> mb_albumid: 26c0f5c2-ed73-4e94-800b-1b120232d39d >> mb_artistid: c0b2500e-0cef-4130-869d-732b23ed9df5 >> mb_releasegroupid: d7fa22fd-29b5-3c52-8500-bcc1e1b6b035 >> mb_trackid: e32badcb-4bb1-43ed-a7fb-d947c17042b4 >> media: CD >> missing: 0 >> month: 10 >> mood_acoustic: 0.97079282999 >> mood_aggressive: 6.59631405053e-10 >> mood_electronic: 0.460209161043 >> mood_happy: 0.555394291878 >> mood_party: 0.00154080230277 >> mood_relaxed: 0.889849960804 >> mood_sad: 0.512115061283 >> mtime: 1969-12-31 17:00:00 >> original_day: 25 >> original_month: 10 >> original_year: 2002 >> rg_album_gain: 0.0 >> rg_album_peak: 0.0 >> rg_track_gain: -4.2 >> rg_track_peak: 1.0 >> rhythm: VienneseWaltz >> samplerate: 44kHz >> script: Latn >> singleton: False >> title: A Sorta Fairytale >> tonal: 0.486311525106 >> track: 02 >> tracktotal: 18 >> voice_instrumental: instrumental >> year: 2002 >> >> >> >> > -- > You received this message because you are subscribed to the Google Groups > "beets" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <javascript:>. > For more options, visit https://groups.google.com/d/optout. > > -- You received this message because you are subscribed to the Google Groups "beets" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
