Bug / chord labelling error in daily ear training challenges

Hi all,

There is a bug / consistent chord naming error in the daily ear training challenges.

The inverted chord labeling in the results/feedback is incorrect. Occasionally you will hear a chord without the distinct, low bass present. In this case (where the distinct bass disappears) this chord never gets named an inversion - even if it still really IS an inversion. In all the challenges I have done over the last few days, it has never been correctly named an inversion.

An inversion is defined by the LOWEST NOTE in a harmony not being the root. That lowest note need not be a ‘distinct’, or especially low pitch note, or on its own in a bass clef. If you have an F major chord, and the lowest note in the harmony is a C4 (i.e the distinct, stand-alone bass is missing - like in today’s challenge) that chord is still in 2nd Inversion. In such a case however, as in today’s challenge, that chord gets labelled an F chord in root position (i.e no inversion!).

Anyone who believes their inversion are getting marked wrong incorrectly when there is no stand-alone bass, this may be the reason.

Here is a recreation of the advanced dictation entry in Hookpad. (Do not bother trying to recreate this in your Hookpad.)

It is evident that the dictation challenges use the old mixing method, as can be heard in the note releases of the piano samples; moreover, if one recreates the actual solution in Hookpad, the A notes in the harmony track and the F note in the bass track would all be an octave lower. But there has never been any “missing” bass; even in the old mixing algorithm the bass range and the harmony range do not overlap unless one uses several altered scale degrees, which for the sake of simplicity have never shown up in dictation challenges.

However I do agree that advanced challenges become more difficult in general since the “Play melody” and “Play chords” checkboxes are gone (their functionality is replaced by the mixer’s mute buttons, but the mixer is obviously inaccessible in a dictation challenge).

Thanks for the video.

I don’t think the bass is missing - by definition, there must, of course, be a bass in any harmony. If Im hearing accurately, the problem is that the audible bass - whichever note that may be - is not determing in the label given to the chord.

Please listen to the progression using the video and maybe you’ll see what I mean. In the 5th chord in the progression, according to the notes marked out in the visual section, the lowest note in the harmony is an F4. But the lowest note you can hear is a C4, which makes the chord an F in second inversion. However, this gets labelled in the results feedback as an F in root position.

Using my headphones and speakers I cant hear the F bass at that point. Perhaps it is my sound system at fault. If it is there though it’s very subtle for sure. I would be interested to see if yourself or the users can hear an F bass at that point.

The lowest note I can hear is a F3 (with C4 as middle C). If you play the audio in a spectrogram, you should also be able to identify the fundamental frequency of F3 around 170 Hz; there is no “ghost tone” around 131 Hz that causes any apparent inversion of the chord.

It is most certainly a fault in your sound system unless for some reason the old set of piano samples all have indeterminate fundamental frequencies for the vast majority of Hookpad users.