Begin the Movement

Composers: Pop* II
Metamusicians: Paul Bodily, and Dan Ventura, CS Department, BYU, Provo, Utah

Description of the Work

Pop* (pronounced Pop-Star) II, which was featured in the 2017 MuMe concert, is an automated pop lead sheet composer. It uses a modular framework to generate verse-chorus structure, rhyme-scheme, lyrics, harmony, and melody. Pop* creates novel full-length pop songs in lead sheet format with no external input beyond an inspiring set of pop lead sheets. To concretely render compositions, we generate both printed sheet music and MP3 audio recordings. MP3 audio files feature computer-sung lyrics accompanied by synthesized piano and bass comping chords.

Technical Description

Pop* II uses a hierarchical Bayesian program learning model, meaning that the concept of a pop composition is factored into subconcept models such as structure, lyrics, harmony, melodic pitch, and melodic rhythm. These subconcepts are further factored until subconcepts represent simple enough ideas to be approximated using data-driven (conditional) probability distributions. Generation of novel compositions is achieved by combining subconcept values as they are probabilistically sampled from subconcept distributions.

Currently in its second iteration, the system uses probabilistic constrained Markov models to generate sequences for each musical viewpoint. In their traditional form, constrained Markov models allow structure to be imposed on sequential data using unary constraints at any of several sequence positions. We have expanded these models to allow for binary relational constraints which enables the system to impose meaningful patterns of repetition—including motifs, verse-chorus structures, and rhyme schemes—while still sampling composition-length sequences as a single Markov process. The relational constraints for these models are also automatically learned from data. This is done by inferring viewpoint-specific structural repeats from existing lead sheets using a self-alignment technique and then converting this structure into a set of relational constraints. 

The system incorporates added elements of autonomy, inspiration, self-awareness, and framing using semantic analysis and social networking to intelligently choose an inspiring set.

Begin the Movement

Begin the Movement is the first song to be composed by Pop* II and was created using structure learned from Twinkle, Twinkle, Little Star; harmonic and melodic models from Over the Rainbow; and a lyric model from Imagine and Hey, Jude.

Audio recording
Lead sheet

Biography

Paul Bodily is a PhD candidate in the CS department at Brigham Young University (BYU). Under the advisement of Dr. Dan Ventura, his research focuses on machine learning in pop music with the intent of building data-driven generative systems.

Dr. Dan Ventura is a CS professor at BYU whose focus is on computational creativity systems generally. Students under his advisement have published systems in domains such as artistic image generation (DARCI), recipe generation (PIERRE), jazz lead sheet composition (CARL), and neology (Nehovah).

Knowing Him

With Ben leaving, it’s caused me to ponder a lot about what makes him such a good friend. Ultimately I concluded that what I appreciate the most about Ben is the Christlike example that he sets. The more I get to know him, the more I feel I get to know Him.

In fact it’s remarkable how much this has triggered a change in how I am approaching coming to know Jesus Christ. My whole life Christ has been “untouchable,” a perfect, yet remote figure. Like, say, Mitt Romney or maybe even President Hinckley. People that I know because of what I hear about them, what I read about them, what I see of them on TV. But Ben—I know Ben. I’ve worked with Ben, I’ve played with Ben, I’ve listened to Ben, I’ve asked Ben tough questions, I’ve seen Ben in tough situations, I’ve seen him in situations where it would be tempting to get frustrated, to say something negative, or to give up—and I’ve seen him do the right thing. And I’ve experienced a feeling each time I watch that, a feeling that that’s the way I want to be, that that’s the way we all need to be to be truly happy.

So at what point does the way I feel toward Christ change from being the way I see and feel toward Mitt Romney to being the way I see and feel toward Ben? Because Ben directly inspires me and Mitt Romney—well, somewhat less so. I love Ben. I want to follow his example, I feel the Spirit when I’m around him.

Now I’ve never had the opportunity to meet Jesus Christ. I really don’t know much about him personally. I don’t know what he’d say to a Costco hot dog. I don’t know what kinds of things would make him laugh. I don’t know how he feels about an Elder’s Quorum move. I don’t know how he prefers to get his exercise. I don’t know how he reacts when he’s happy. I don’t know how he reacts when someone says something awkward. Except of course beyond trying to find some scripture that seems to relate to those things. 

Yes, I may have learned a lot about Him, but I don’t really feel like I know him, with the lower-case ‘h’, if that makes sense, and knowing him, at least as far as might be mirrored by my experience having been friends with Ben, makes an enormous difference in how it inspires me to see a man, with foibles and who is “touched with the feeling of our infirmities,” “in all points tempted like as we are.” I need to see the ways that He is like me, both for His successes and for His struggles, to see that my potential is somehow related to His. To see that the greatness I see in Him comes in ways that I am able and inspired to follow. I guess, as Lehi puts it, to see the “opposition in all things,” because without that element, without that attribute, as Lehi continues, “righteousness could not be brought to pass, neither wickedness, neither holiness nor misery, neither good nor bad.” Picturing a Christ who “cannot be touched” or is even in some degree less touched is to picture a Christ who “[has] no joy, for [he knows] no misery; doing no good, for [he knows] no sin.”

Since coming to that realization, I’ve thought a lot more about Christ. I even catch myself thinking about him while watching random TV shows. I’ve tried to picture the “him” behind the “Him”. I’ve read the scriptures about Christ more carefully, picturing a man “touched” and “tempted” and “yet without sin”. I’ve found myself pondering who He really was. I’ve found richer meaning in simple phrases that I’ve read hundreds of times like, “then Jesus was led up of the Spirit, into the wilderness, to be with God,” and “this is life eternal, that they might know thee the only true God, and Jesus Christ, whom thou hast sent.”

I’ve never understand exactly what it meant to know Christ. I’ve always known that I should want to know Him. But now I really want to know him. My friend, Ben, taught me more than I think he’ll ever know about the Savior, because he reflected something that directly emanated from the Savior, something that caught my eye and touched my heart. I realized that beyond the love for my friend, I had a love for the goodness, for the light, for the truth he bore. Recognizing that what I saw in my friend was but a reflection from somewhere or someone else made me realize that the source of all goodness is a real, tangible source. It made me realize that were I to come to know Christ as I have come to know my friend, those feelings of love, light, goodness, truth, and inspiration would be deeper than perhaps I ever thought imaginable. This is not just some man who changed the world. This is not just some person my Sunday school teachers tell me I should want to know better. He's not just a name in a book. He is a “treasure hid in a field; the which when a man hath found, he hideth, and *for joy* thereof goeth and selleth all that he hath, and buyeth that field.”

Pop* and Lyrist Debut

Description of the Work

Pop* (pronounced Pop-Star) is an automated pop lead sheet generator. It uses a modular framework to generate verse-chorus structure, rhyme-scheme, lyrics, harmony, and melody. The lyrical module, which we call Lyrist, functions as a stand-alone lyric generation system. Pop* and Lyrist work in concert to create novel full-length pop songs in lead sheet format on their own with no external input (beyond an inspiring set of pop music and lyrics). To concretely render compositions, we generate both printed sheet music and MP3 audio recordings. MP3 audio files feature computer-sung lyrics accompanied by synthesized piano and bass comping chords.

Technical Description

Pop* uses a hierarchical Bayesian program learning model, meaning that the concept of a pop composition is factored into subconcept models such as verse-chorus structure, rhyme-scheme, lyrics, harmony, and melody. These subconcepts are further factored until subconcepts represent simple enough ideas to be approximated using data-driven (conditional) probability distributions. Generation of novel compositions is achieved by combining subconcept values as they are probabilistically sampled from subconcept distributions. 

Lyrist represents a subconcept model of lyrics (conditioned on melody and intention). Pop* creates lyric templates by recombining lyrical phrases from existing songs. These templates serve as input to Lyrist which intelligently replaces words in the template to create new lyrics. Lyrist uses word embedding (i.e, a numeric vector which represents the semantic meaning of a word), vector operations, and constraint (e.g., rhyming and part-of-speech) filters to generate novel lyrics that evoke an intended theme and rhyme structure.

More complete descriptions of these systems will be found in the proceedings of ICCC 2017 and MUME 2017.

Best Song So Far (BSSF)

Here is a recording of our "Best Song So Far" with music generated by Pop* and lyrics created by Lyrist:

The sheetmusic can be found here.

Biography

Paul Bodily is a PhD candidate in the CS department at Brigham Young University (BYU). Under the advisement of Dr. Dan Ventura, his research focuses on machine learning in pop music with the intent of building data-driven generative systems.

Ben Bay is an undergraduate research assistant pursuing his B.S. in CS at BYU. Under the advisement of Dr. Dan Ventura and mentorship of Paul Bodily, his research focus is on systems that generate lyrics.

Dr. Dan Ventura is a CS professor at BYU whose focus is on computational creativity systems generally. Students under his advisement have published systems in domains such as artistic image generation (DARCI), recipe generation (PIERRE), jazz lead sheet composition (CARL), and neology (Nehovah).

'Tis a Gift

Larry Howell, a professor in Mechanical Engineering at BYU, recently spoke at a graduate student symposium on the topic of Faith & Scholarship. In light of his renown and success as an academic, I was struck by something he said (and I paraphrase): "At the final judgment, God will not be interested in how many degrees we earned or how many publications we had, but He will ask what we've done with our gifts."

"To every man is given a gift by the Spirit of God," we read in Doctrine and Covenants 46:11. It seems apparent that we each have different strengths and talents. Discovering our gifts requires more than simply identifying what we're gifted at (though it may be a good place to start); gifts require passion.

There are certainly some gifts that we might not be particularly passionate about, but which we should seek for, hence the admonition to "seek ye earnestly the best gifts" (Doctrine and Covenants 46:8). Some of these might include parenthood, charity, and faith. Some gifts are essential to developing the character of Christ.

But just as each of my children have unique gifts and talents, each of us, as literal children of heavenly parents, are blessed with unique gifts and talents, passions for different areas of human endeavor that would drive us not only to make the world in that sphere a better place, but also a predisposition towards seeking and receiving inspiration in those areas.

It's common, in my opinion, for people to go through life without ever discovering or magnifying the gifts they've been given. But those who do become a blessing to any and all with whom they come in contact.

I can readily picture God at the final judgment taking a very personal moment to show us what has been so unique about us, even since the premortal world and then asking us how we used those talents to bless others.

In picturing that moment, I suspect He will also ask us about one particularly important gift that we are to receive in mortality, which is the Gift of the Holy Ghost. A gift to guide and direct us to find other gifts. "How did you use that gift?" He will most assuredly ask. That's probably not a question I think about enough.

Primary Singing Time: Combo Lock Puzzle

Here's an idea to get Primary kids excited during Singing Time.

Materials needed:

• A chalkboard and chalk
• A combination lock (optional)
• A prize (e.g. bag of dum-dums)

In preparation for this singing time activity, draw something like the following on the chalkboard:The numbers around the edges are chosen at random between 1 and 100, but must include the numbers in the code for the combination lock.

In order to solve this puzzle the children will need to sing songs which will stepwise reveal the traversal of the puzzle and the solution to the combination lock. The first number in the combination, together with the first direction, is already given (down-right), and indicates the song that must be sung in order to progress. In this example, the children must sing "Praise to the Man" (to the chorister's satisfaction) in order to earn the next arrow. The chorister may choose to test how well the children know a particular song before they decide that the arrow is earned.

Once the song has been sung, the chorister gives the next arrow, which in our example is again down-right:The children must next sing "Search, Ponder, and Pray" in order to progress. And so on.

After every third or fourth song, the arrow should lead to the next number in the code for the combo lock. Thus, after 6 or 7 songs, they will have earned all three numbers to open the lock:When the arrow points to a number, the chorister may let a child choose the next song or may elect to simply give the next arrow gratis.

Though the combo lock and prize could technically be optional, the enthusiasm inherent in the activity depends on the presence of a reward. Even just a bag of dum-dums should be sufficient to make the activity a raging success. Alternatively, you could have a member of the primary presidency act as the "combo lock" and will only give the prize upon receiving the correct combination.

If you'd rather not have to draw everything up on the chalkboard the activity could be modified to be more like a scavenger hunt around the room with clues hidden bearing the name of the song and the numbers for the combination lock. Again, the chorister would simply give the directions of where to find the next clue. The challenge here would be keeping the clues with the numbers for the lock in order.

One nice thing about this activity is that to fill time you can make up the puzzle traversal on the fly (don't tell the kids!) to include more or less songs. You could even repeat a song in the traversal if you are working on a song for an upcoming sacrament program. If you have ideas for variations on this activity, please let me know! Have fun!

No More "Mr. Trusting User"

According to the 2006 Internet Crime Report from the FBI Internet Crime Complaint Center, auction fraud and non-delivery are the cause of nearly sixty-five percent of all internet crime complaints. No other category of complaint amounts to even five of the remaining thirty-five percent. What jumps out to me is that these two categories of internet fraud do not represent technological cunning and manipulation but merely a regrettably faulty reliance on a stranger’s discretion.

So often we picture the perpetrator as one who takes advantage of the anonymity of the internet or one who uses a superior understanding of computers and programming to achieve an immoral purpose. However, these statistics tell us that sixty-five percent of those “bad guys” aren’t hiding and aren’t particularly cunning; they’re just dishonest. And as relatively little has been done as far as regulatory efforts, these criminals are largely successful.

This reflects how difficult it is to adapt to the very new and different world of the internet. People are used to dealing with people face to face or at least in a way that leaves a physical path for crime fighters to follow. With the internet, we’ve been slow to recognize and utilize what paths exist (many times for political reasons) and thus many mischievous malefactors (who may be more cautious and less present in “offline” markets) are making their debut in a realm that has yet to learn how to properly raise or deal with a red flag.

Perhaps someday we will each have a universal account; an account with our global identity, all of our financial assets, all of our credit and credentials. Then the consequences of bad behavior online won’t be so easy to evade or ignore. But until that day, our motto must be, “be safe or be sorry”. We can’t assume anything but the worst about an online newcomer. Such an attitude may put a damper on business and commerce, but these negative effects will have been worth it if they causes us to recognize the need to focus our attention as much on securing the internet as on building it.

An Ounce of Prevention

My father raised us on the maxim, “an ounce of prevention is worth a pound of cure”. Though few will disagree with this proverb, we’re smothering the greatest concerns of our day with cure and leaving the bottle of prevention untouched on the shelf. I’m talking about internet safety.

My greatest childhood memories were learning to do dangerous things in a safe way. When my dad taught me to mow the lawn, he made sure that I understood not to put my hands or feet down near the blades. When he taught me how to use the edger, I had to wear protective eye-wear. When he taught me how to drive a motorboat, I learned to prop up the engine before pulling the boat out. Our most entertaining family nights were when we pulled out the old plastic rope ladder and practiced our back-up fire plan of climbing out the second-story window.

Unfortunately, despite parents best hopes to protect their children, some of the greatest dangers, those of the internet, go largely unaddressed. We put the computer in an public place, we install filters, we check browser histories, but these things are all cures, not preventions. It would be like taking the time to program 9-1-1 into your speed dial for when your child chops his hand up in the lawn mower rather than taking the time to warn of the danger itself.

Don’t get me wrong, cures are important. But prevention is better. Learning about the internet and teaching children of its dangers is the most important practice we can employ to steer clear of online mishaps. Before reaching for the bottle of cure, let’s use a little more prevention.