1
$\begingroup$

Musical notes

Musical notes videos

Piano

Can AI, Machine learning, Data science, Computer vision, image processing technologies assist in interpreting musical notes ?

Input dataset : Musical notes

Output : Sound played.

A python program which can assist in interpreting or converting text to speech ?.

The text inputted is Musical notes captured as a image by camera as input device.

Improve this question
$\endgroup$
4
  • 2
    $\begingroup$ What are your findings so far? Are you back to our site to pollute it with poor questions? Please, don't do that. For the 1 millionth time, please, read ai.stackexchange.com/help/how-to-ask and ai.stackexchange.com/help/on-topic. To ask a good question, you need to do a little bit of research before asking it and share with us your findings. You can easily find answers to this question on the web. There's no excuse. If you could find our site to ask this question, you could also have find the answer on the web. $\endgroup$
    – nbro
    Commented Mar 17, 2022 at 9:28
  • 2
    $\begingroup$ Also, why are you linking us to images of notes? Do you think that people here don't know what a musical note is? It makes no sense to me. $\endgroup$
    – nbro
    Commented Mar 17, 2022 at 9:33
  • 2
    $\begingroup$ Similar if not duplicate question ai.stackexchange.com/q/11431/2444. $\endgroup$
    – nbro
    Commented Mar 17, 2022 at 10:57
  • 1
    $\begingroup$ I am closing this question for being too broad, because it is basically asking about an entire pipeline of a rather big task (image inputs --> sound outputs based on image contents) whether "it is possible using AI/ML". This invites either just "yes"/"no" answers, or 50-page reports on how to handle the entire pipeline. We require more specific and precise questions about well-defined, specific problems. $\endgroup$
    – Dennis Soemers ♦
    Commented Mar 17, 2022 at 15:36

2 Answers 2

Reset to default
4
$\begingroup$

AI/ML can solve the task described, a solution is as below:

  • Regular image processing algorithm (pixel row with min black pixels, adjacent rows are considered as 1) to split the sheet music (as image) into rows (red lines)
    enter image description here
  • Regular image processing algorithm (pixel column with min black pixels, adjacent columns are considered as 1) to split a row into notes based on scanning vertically (blue lines)
    enter image description here
  • Do image classification on every cell created by red lines and blue lines to get notes, and play them
  • Real-time image classification may be slow and not possible to play note by note directly, the programme should process the whole sheet music first.
Improve this answer
$\endgroup$
2
$\begingroup$

LMGTFY ;)

The problem is called Optical Music Recognition. Here you can find tutorial that desribes OMR with deep learning and here you have scientific paper. I think it is very good start for your investigations.

Improve this answer
$\endgroup$

Not the answer you're looking for? Browse other questions tagged .