AI Accuracy: Medium
Using audio data models, the Keytone AI has predicted the most suitable note arrangement and chords for Grace based on publicly available data.
The AI accuracy level for Grace is about medium in accuracy confidence. This is determined by a healthy dataset which has a good correlation of notes and how they are arranged in the relation to common music modes and other music theory factors.
How did we do?
The system is always receiving new data and tweaks here and there are made for maximum quality. Feedback is always important so please feel free to let us know!
Our song key finder algorithm has predicted that the key signature for "Grace" by Jeff Buckley is most likely E Minor (Relative key G Major ). Read More
Besides checking what notes are heard in the song, it also checks for repetition and patterns associated with chords and scales. It also detects the best suited root note so further predict the mode.
The sample set of notes can determine how likely of a correct decision is made. If the system is unable to detect many notes, then it is least likely to make a likely prediction. Some of these factors are true of professional musicians too but humans can 'feel' correctness through ear training. Too many notes on the other hand is equally difficult and will result in too many possible predictions. At which case, the system will then average out for the more likely result.
Ironically the human element can be the biggest impact on the accuracy! Musicians generally stick to the music theory but there's quite often notes misplaced or added based on artistic principles. Perhaps the additional notes can unexpectedly change cadence to the song or the entire key signature could change. Once again, professional musicians can ignore these elements when working out how to play a song, something that AI can quite often misunderstand.
The system has detected the following notes in Grace: D♯/E♭, A, B, G, D, A♯/B♭, F, C & F♯/G♭, the most common note being E.
The notes discovered do not apply to singular instruments but all detected notes. Each count of note determines the likelihood of the closest signature match along with best chord based on common and repeated patterns throughout the song.
Below shows all the discovered notes with the song “Grace” by Jeff Buckley by the Keytone AI. This data is used to determine the closet matching mode which makes the chord matching process refined.
Below shows the full range of notes discovered and key number equivalent with octaves. This incorporates every instrument and vocal notes discovered within “Grace” by the Keytone AI. Like the above condensed noteset, this information can be used to find the best pitch for chords and modes.
Below shows the AI using the notes provided to predict the closest key signature. If there is not enough notes to correctly identify the key signature, the system will re-organise the notes then re-attempt to suggest the best match.