Share this project

Done

Share this project

Done
I will make several pre-trained Google Tensorflow Machine Learning datasets available for everyone to produce song snippets.
I will make several pre-trained Google Tensorflow Machine Learning datasets available for everyone to produce song snippets.
3 backers pledged CA$ 30.00 to help bring this project to life.

About this project

Rocksetta: Neural Network, Machine Learning MP3 Creation project video thumbnail
Replay with sound
Play with
sound

CA$ 30.00

3

So I am not going to make a robot that can play the piano, but that does make a nice visual of what I am trying to do.

I wish to make short pieces of music, based on past successful songs by entering many small snippets into a Tensorflow Recurrent Neural Network (RNN). A bit of software from Google that allows computers to learn a skill and reproduce it's own version of that skill. 

Imagine entering every Beatles Generation song snippet and having a computer create it's own version of Beatles style songs, lots of them! Some songs might be too close to the originals, some might be just plain really bad, but occasionally the computer might be able to produce a gem that takes you back to another time. Imagine what a musician could do with that golden rift from the past created by the future.

To give a more modern example: perhaps Michael Bublé or Kate Perry would be interested in what a machine could come up with from their successes.

Please make this decades long wish a reality by donating a few dollars to my project.

The present problem is the song entry. Most Machine Music Learning tools uses huge datasets of thousands of midi generated multi-track files. I have not liked the quality of this approach, and prefer simple melody files. Uploading high quality melody files has been a problem. I would have to hire a musician to check the quality of the present database and to upload more choice snippets. (If anyone wants to painstakingly enter hundreds of songs snippets for free just for the love of music research, I would be very impressed)

Another cost will be the development platform which is presently $19.00 USD per month. To have this site available for a few years will cost ($19.00 x 24)  ~$456.00 USD

Most of the difficult programming has been done over the last decade and can be viewed on the http://rocksetta.com website. The Machine Learning part is reasonably new but can be viewed on various of my github sites at  https://github.com/hpssjellis/char-rnn-tensorflow-music-3dprinting and several others at https://github.com/hpssjellis?tab=repositories  

An example machine learning music making but demo is working at https://big-char-rnn01-rocksetta.c9users.io/rnn-both.php

A strange coincidence is, that the music notation I have developed over the last decade ( http://keyfreemusic.com ) actually works very well with machine learning since every combination of characters produce sensible musical sounds. (I have developed four styles of music notation: Keyfreemusic, Music Letters, Rocksetta Notation for database entry and the best a readable diagram notation.)

The final goal is to make a Web App that allows anyone to generate genre based song snippets from several datasets optimized for different song styles. This gives people a chance to experience some of the benefits of machine learning as an assistance to creativity. It is then up to musically creative people to take the song snippets and make full fledged songs.

Please support this project and lets see what music we can make!

Risks and challenges

The three main risks are:

1. Updates to the Google Tensorflow Machine Learning software. Sometimes updates can cause issues with the present software. A solution is to stay with an older version of Tensorflow, or update the Machine Learning code.

2. Garbage in creates garbage out: The quality of the entered song snippets influences the output. Much of my costs will be to pay musicians to correct and enter high quality song melody snippets.

3. Not enough songs to generate a quality output. Still working on techniques to improve the quality of the output using minimal input. A present technique is to randomize the entry of the songs several times in the dataset. Other ideas are to populate a main dataset of successful songs with the new songs to study.

A challenge is to keep the song snippets short enough and as a single note melody so as to not get in trouble with the original Artists. Obviously if there are issues, those song snippets will be removed from the database.

Another challenge will be to upgrade the main Adobe Flash Web App to HTML5 Web Audio for modern browsers. This is not an immediate issue and is something I will enjoy doing as time permits.

Learn about accountability on Kickstarter

Questions about this project? Check out the FAQ

Support this project

  1. Select this reward

    Pledge CA$ 1 or more About $0.75

    Good Karma

    I will post your first name as a "Thank You" on my Kickstarter Good Karma webpage at http://rocksetta.com/kickstarter/goodkarma.html

    Less
    Estimated delivery
    1 backer
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  2. Select this reward

    Pledge CA$ 2 or more About $1.50

    Good Karma +

    I will post your first name or the name of someone you choose as a gift on my "Thank You" Kickstarter Good Karma webpage at http://rocksetta.com/kickstarter/goodkarma.html

    The "Thank You" will include a link to a unique downloadable MP3 of a randomly generated song snippet created by the best machine dataset to date

    Less
    Estimated delivery
    0 backers
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  3. Select this reward

    Pledge CA$ 3 or more About $2

    Helper Musician

    Record a quality song snippet at http://rocksetta.com and I will send you a machine generated MP3 of several song snippets.

    This is not easy to do and takes some determination and skill. There are several approaches.

    On the http://rocksetta.com site you will have access to a midi version of your song and the sheet music for your song, as well as other musical visual representations.

    Your song may become part of the main song dataset used for the base machine learning experiments.

    Less
    Estimated delivery
    0 backers
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  4. Select this reward

    Pledge CA$ 5 or more About $4

    Early Bird Song Snippet

    I will email a unique Machine Generated song snippet in MP3 format from the present machine trained dataset

    Less
    Estimated delivery
    Limited 0 backers
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  5. Select this reward

    Pledge CA$ 8 or more About $6

    Machine Generated Song Snippets

    I will email about 8 short unique machine generated song snippets in MP3 format from the present machine trained dataset

    Less
    Estimated delivery
    Limited 0 backers
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  6. Select this reward

    Pledge CA$ 15 or more About $11

    Suggest a Song

    Suggest a song for us to upload a short snippet of it to the Rocksetta database to be used in future machine learning datasets.

    You get an emailed MP3 selection of machine generated song snippets created in part using your song suggestion.

    Less
    Estimated delivery
    0 backers
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  7. Select this reward

    Pledge CA$ 20 or more About $15

    Music Research Philanthropist

    In appreciation of your covering my Cloud Development platform single month cost.

    I will email many short machine generated song snippets in MP3 format from the best dataset made before June 2017

    Less
    Estimated delivery
    Limited 1 backer
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  8. Select this reward

    Pledge CA$ 120 or more About $90

    Suggest a Genre

    Suggest a music genera and supply up to 10 song titles you would like to be uploaded to the Rocksetta database.

    Receive an email MP3 of as many song snippets as you would like using your songs merged with an appropriate song dataset.

    Less
    Estimated delivery
    Limited 0 backers
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  9. Select this reward

    Pledge CA$ 400 or more About $300

    Suggest an Artist

    Suggest an Artist along with up to 30 song titles to have song snippets entered into the Rocksetta database.

    Receive an email MP3 of as many song snippets as you would like generated using your Artists songs merged with an appropriate song dataset.

    Less
    Estimated delivery
    Limited 0 backers
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  10. Select this reward

    Pledge CA$ 4,000 or more About $3,002

    Meet and Greet or Presentation

    I will present our findings to your group anywhere in the US or Canada. You cover the plane cost and accommodations (Potentially for 2 people).

    I will give a 2 hour presentation to any number of people talking about: My experiences, frustrations, history and joy working with Machine Learning, Music, Robotics and 3D Printing.

    I reserve the right at any time to refuse this invitation.

    Less
    Estimated delivery
    Limited 0 backers
    CA$
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.

Funding period

- (23 days)