Open Amarelen opened 1 year ago
Unfortunately, we don't have the codes for processing MIDI to our data representation, although we also hope to have it. Also, to construct another aligned melody-lyrics dataset, lyrics with time information are needed in addition to MIDI files.
On Thu, Aug 17, 2023 at 3:43 PM Amarelen @.***> wrote:
Sorry to bother. Does anybody know if I want to change the data source to my own midi files, what form the data should be prepared and which code to use? Thanks!!!
— Reply to this email directly, view it on GitHub https://github.com/yy1lab/Lyrics-Conditioned-Neural-Melody-Generation/issues/11, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMQGHTF3GFL6NUOUAULJ5TTXVW4SLANCNFSM6AAAAAA3TSXG2I . You are receiving this because you are subscribed to this thread.Message ID: @.*** com>
Unfortunately, we don't have the codes for processing MIDI to our data representation, although we also hope to have it. Also, to construct another aligned melody-lyrics dataset, lyrics with time information are needed in addition to MIDI files.
Unfortunately, we don't have the codes for processing MIDI to our data representation, although we also hope to have it. Also, to construct another aligned melody-lyrics dataset, lyrics with time information are needed in addition to MIDI files.
I'm really grateful for your quick and sincere reply! I noticed in the original LMD-full dataset there are only raw midis, and in the "LYRICS-MELODY DATA ACQUIREMENT" part of the paper you have mentioned the parsing method, which I think is really brilliant. I wonder what's the code you used to turn the raw midi data and its paired lyric to the "Sentence_and_Word_Parsing" folder(in which are all .npy files) in the provided “lmd-full_MIDI_dataset”. Really sorry if I didn't explain clearly enough in the last question and if you can provide this part of the code or its access, I would greatly appreciate it.
Unfortunately, we don't have the codes for processing MIDI to our data representation, although we also hope to have it. Also, to construct another aligned melody-lyrics dataset, lyrics with time information are needed in addition to MIDI files.
I'm really grateful for your quick and sincere reply! I noticed in the original LMD-full dataset there are only raw midis, and in the "LYRICS-MELODY DATA ACQUIREMENT" part of the paper you have mentioned the parsing method, which I think is really brilliant. I wonder what's the code you used to turn the raw midi data and its paired lyric to the "Sentence_and_Word_Parsing" folder(in which are all .npy files) in the provided “lmd-full_MIDI_dataset”. Really sorry if I didn't explain clearly enough in the last question and if you can provide this part of the code or its access, I would greatly appreciate it.
Sorry to bother, have you solved this problem yet?
Unfortunately, we don't have the codes for processing MIDI to our data representation, although we also hope to have it. Also, to construct another aligned melody-lyrics dataset, lyrics with time information are needed in addition to MIDI files.
I'm really grateful for your quick and sincere reply! I noticed in the original LMD-full dataset there are only raw midis, and in the "LYRICS-MELODY DATA ACQUIREMENT" part of the paper you have mentioned the parsing method, which I think is really brilliant. I wonder what's the code you used to turn the raw midi data and its paired lyric to the "Sentence_and_Word_Parsing" folder(in which are all .npy files) in the provided “lmd-full_MIDI_dataset”. Really sorry if I didn't explain clearly enough in the last question and if you can provide this part of the code or its access, I would greatly appreciate it.
Hello, I am also interested about whether you have found a solution to this problem? Many thanks.
Sorry to bother. Does anybody know if I want to change the data source to my own midi files, what form the data should be prepared and which code to use? Thanks!!!