Team Contract ​

Members Oreoluwa Alao, Promia Chowdhury, Dima Yanovsky, Joseph Ye

Expected level of effort ​

• For this group assignment, we will all aim to contribute equally to our project and develop a mutual, supportive environment in doing so. Ideally, we want to receive an A, and expect each member to complete their weekly objectives and deadlines, as they correspond to the official class timeline. If a member is unable to do so—in light of difficult circumstances—, we will attempt to reduce objective scopes, and/or cooperate together with 6.104 staff to receive accommodations, as needed.

Personal goals ​

• Promi: My goal is to create and design a software with my team that I can be proud of, that I contributed a fair and significant portion of effort into 😃

• Joseph: My goal is also to create and design a full-stack application that will have a positive impact on the world.

• Ore: First priority: usability, adherence to VSD and application robustness; then the perfection of algorithm

• Dima: Develop a robust application that will serve as a proof-of-concept of something that can have a real positive impact on the community.

Frequency, length, and location of team meetings ​

• Will have weekly or bi-weekly team meetings during weeks when we don’t have lectures to work on our project. Length will be as needed to complete agenda items. Location will be wherever we decide upon. If organized on short notice, we will meet virtually on Zoom.

Quality of Work ​

• After each person has contributed their part to the project, we will have a meeting to go over the entirety of the app to assess the quality.

• Decisions will be made if most group members agree to it. Disagreements will be resolved through assessing the pros and cons of ideas and through a majority vote.

Tasks ​

• Corresponding to the class assignment deadlines, at the beginning of each week, we will assign a significant milestone to each member, with the intention of coalescing and reviewing each other’s work when it is time to submit. We will assign based on outlining milestones that are sub-problems of the week’s assignment.

• If deadlines are not met, we expect to notify each other immediately and cooperate on reducing the week’s scope, if needed. We will also make sincere attempts to communicate any problem to our TAs.

• A member’s task will typically be a major programming milestone (such a concept implementation) or a collection of smaller tasks.

Decisions ​

• Decisions will be made after reviewing options with every team member. In the case of disagreements, we will make a decision based on the majority opinion, or consult an instructor or a TA.

Draft Impact Case ​

Make memorization through song more accessible to students by automating the songwriting process: Songs are effective mnemonic devices. The Broadway play Hamilton made it easy for the biography of Alexander Hamilton to be memorized by a casual audience without requiring extensive study. However, due to the amount of time it takes to create memorable lyrics, it is not feasible for students to create them when studying school notes. An app that automates the songwriting process for school notes could let students benefit from songs as a study tool, helping them spend less time studying.

Interviewee Roles ​

Background Research ​

Problem and Justification ​

Problem: The challenge of hard-wired memorization in academics poses a significant obstacle, especially for students grappling with neurodivergence or learning disabilities. This difficulty in independent retention of academic content creates a considerable barrier to effective learning. Many students find it arduous to commit information to memory without additional study aids, impeding their progress and long-term success in their studies. To address this issue, we propose the development of an intuitive and user-friendly song memorization app. This solution aims to empower students by providing them with a tool to effortlessly generate mnemonic songs tailored to their learning needs, thereby enhancing their ability to retain academic material independently. Research has shown that students, particularly those with neurodivergent traits or learning disabilities, face substantial difficulties in independently retaining academic content. This challenge is highlighted in the study conducted by Kanatovna et al., titled "An Innovative Approach to Memorizing Words Via Modern Songs" [Kanatovna]. The research focused on investigating the impact of songs as mnemonic devices on vocabulary retention during the language learning process.

The study's preliminary literature review revealed a wealth of existing data that underscore the benefits of incorporating songs and other musical devices into the learning process. These musical elements were found to project lexical and phonic patterns onto new information that would otherwise be challenging to remember or distinguish. This finding emphasizes the potential of music as a powerful tool for memory enhancement.

Furthermore, the study conducted by Kanatovna et al. demonstrated a positive trend in using songs for rote memorization of new idioms and vocabulary among grade-level students learning a new language. This reinforces the notion that songs serve as effective mnemonic aids, particularly in language acquisition, where the memorization of vocabulary is crucial.

The wealth of evidence provided by this research supports our belief in the efficacy of utilizing songs as mnemonic devices through our proposed app. By making this powerful learning tool accessible to a wider audience, we aim to address the pressing issue of independent memorization faced by students, particularly those with neurodivergent traits or learning disabilities.

Existing Solutions To Our Problem ​

ChatGPT: ​

An algorithm for generating song lyrics

Pros:

• Can generate lyrics about inputted text

Gaps:

• The generated lyrics have no accompanying tune, and are thus unsingable out of the box.
• To constrain the generated lyrics from ChatGPT to an existing tune, prompt engineering is required

SongPad: ​

User-friendly editor for collaborative song-writing with AI, providing tools like synonym finder, and a song line generator. Allows sharing songs.

Pros:

• Supports Collaborative writing of multiple users
• Gives user control over the lyrics in the song
• Song outline to guide your ideas
• Has Blogs to share tips on songwriting

Gaps:

• No tune (user is required to come up with one, and be familiar with Chords)
• No guarantee that resulting song is memorable

Hamilton: Soundtrack informs about early American history ​

• Pro: Memorable (as shown by its popularity)
• Gap: Cannot be reproduced in easily for other subjects: written by expert humans over years

Flocabulary ​

Flocabulary is an educational platform that offers a library of songs, videos, and activities for K-12 online learning.

Customization: Flocabulary provides pre-made content, which may not align with specific curriculum requirements or personal learning needs. Gap: Not an app that generates songs based on user-inputted notes or textbooks that would offer personalized learning.

Interactivity: Flocabulary has interactive elements but may not fully engage users in the songwriting process. Gap: Not an app that lets users be part of the song creation, whcih could enhance engagement and retention.

Diverse Learning Strategies: Flocabulary focuses on hip-hop to teach concepts which may not cater to all learning preferences. Gap: Not an app offering a variety of musical genres which could cater to a wider range of learners.

Content Complexity: The complexity of Flocabulary's songs may not match the required academic depth for older students or for advanced subjects. Gap: Not an app with adjustable complexity levels that could adapt to different educational stages.

LingoClip ​

LingoClip is a language learning tool using music to enhance vocabulary, grammar, and pronunciation.

Learning Focus: LingoClip is designed for language learning and does not address other educational content needs. Gap: Not an app that translates a wide array of academic materials into songs could offer a broader educational tool.

Diverse Learning Strategies: LingoClip uses music as a memorization tool but is limited to language learning. Gap: Not an app that could expand the strategy to other subjects, offering memorization beyond languages.

Interactivity: While LingoClip includes interactive lyrics and game modes (which are nicely done), it doesn’t involve users in the songwriting process. Gap: Not allowing students to create songs from their study notes, which could increase interactivity and personal engagement.

VSD Analysis ​

Indirect Stakeholders:

Potential Impact: Because the nature of our app is one of “generation,” it is by default that our first ideas already roamed into AI generated solutions. However, we must use such technology carefully, as many artists and musicians object to their voice and likeness being used in AI model training and software. This is a major point of contention for SAG-AFTRA.

Mitigation: To avoid copyright and ethical issues regarding potential artificial sound generations, we will only utilize assets already belonging to the public domain.

Privacy and Security (Information Ownership):

Potential Impact: Users may be concerned about the privacy of their notes and potentially sensitive study materials when using the app.

Mitigation: Implement strong encryption and secure user authentication to protect user data. Clearly communicate the app's privacy measures and emphasize user ownership of their information.

Variation in Human Ability (Sight, Hearing, Touch, Speech, Motor Control):

Potential Impact: Users with varying physical abilities may encounter challenges in interacting with the app, especially if it relies heavily on audio elements.

Mitigation: Design the app to be accessible for users with different abilities. Provide alternative options for interaction (e.g., text-based input) and ensure compatibility with assistive technologies.

Crossing National Boundaries:

Potential Impact: Different countries have different rules, customs, and infrastructures that may affect the use of the app. Users in different countries may have varying experiences or encounter unique challenges.

Mitigation: Conduct thorough research on the legal and cultural contexts of different countries where the app will be used. Ensure compliance with local regulations and adapt features or content to align with specific cultural norms or legal requirements. Provide clear guidance for users in different regions to navigate any country-specific considerations.

Long-Term Health and Well-Being:

Potential Impact: Daily interactions with the app could have long-term effects on users' mental and emotional well-being, especially if they become overly reliant on the app for studying.

Mitigation: Encourage balanced study habits and provide resources for effective learning beyond the app. Include features that promote healthy study practices, such as breaks and time management tools.

Social Impact:

Potential Impact: The app may have social implications, influencing how users interact, communicate, and form relationships with others. This could lead to both positive and negative effects on social dynamics.

Mitigation: Conduct user research to understand potential social impacts and incorporate features that promote positive social interactions. Provide tools or guidelines to encourage constructive communication and collaboration among users. Monitor for any unintended negative social consequences and be prepared to address them promptly.

Non-targeted Use:

• Impact: the app could be used to write songs about topics that are not educational
• Mitigation: user input will be filtered against non-educational content

Direct stakeholders:

Impact:

• Since the app is for educational use by students, the content produced by the app must be age-appropriate (no foul language, no inappropriate suggestions)

Mitigation:

• User-delegated Moderation will be a core feature of the app for controlling user
• The social features of the app can revolve around classroom structure of authority to ensure the app is used for correct purposes (like google classroom requires a teacher, Kahoot require)
• Song lyrics, before being shown to users or published in the app, will be vetted for the appropriateness of topics

Indirect stakeholders:

Impact:

• Schools might not want notes from their curriculum posted online (i.e. not have an open-source education policy)
• The use of existing tunes by must respect the IP of songwriters; we must make Sure that copyright is respected despite the fact that parodies are encouraged (and regularly generated) on the app

Mitigation:

• Avoiding the publishing of songs that violate the 4-factor test for Fair Use; generated songs will not be revealed to users if it violates the 4-factor test, namely: the purpose must be educational (for learning school notes) and the topic must be substantially different from the song that is being parodied

Interview Summaries ​

Interview #1 with Wilson ​

Wilson provided valuable insights into his experiences and perspectives on using mnemonic devices, particularly songs, for studying. He shared that he employs acronyms as a mnemonic device to memorize biological hierarchies, emphasizing their effectiveness. However, he also acknowledged a potential drawback - finding songs somewhat bothersome due to their tendency to linger in one's memory. Despite this, he recognized their significant contribution to memory retention. Furthermore, Wilson highlighted the challenges he faces when attempting to create effective mnemonic songs. He pointed out the difficulty of composing a song and expressed the belief that an app could streamline this process. Drawing a comparison to the efficiency gained from using pre-made flashcards, Wilson emphasized the potential time and effort savings offered by such a tool.

While Wilson did not have specific experiences using mnemonic songs for studying, he shared an example from his work in creating wedding videos. He described the intricacy of altering song lyrics to fit a specific theme while maintaining the original rhythm, underscoring the time-consuming nature of this process. Identifying two primary obstacles, Wilson outlined the importance of syllable alignment and narrative coherence when creating memorable lyrics. He stressed the need to ensure that the number of syllables aligns with the song's rhythm and highlighted the significance of constructing a cohesive narrative to enhance memory retention.

In terms of effective mnemonic songs, Wilson emphasized the importance of a catchy chorus encapsulating key information for retention. He suggested that the app should recommend songs with strong choruses and have the capability to vocalize them.

Looking ahead, Wilson anticipated that an app automating the songwriting process for school notes would revolutionize the studying experience. He envisioned increased enjoyment in studying and enhanced memorization capabilities. Additionally, he foresaw potential for collaborative learning, envisioning scenarios where friends could come together to sing. Wilson also highlighted the opportunity for students to exercise creativity, allowing them to customize lyrics for improved flow and originality. While Wilson did not express major concerns, he briefly mentioned potential copyright considerations as a minor consideration.

In summary, Wilson's feedback provides a comprehensive understanding of his experiences and expectations regarding the proposed app. His insights underscore the potential benefits and impact it could have on the studying habits and experiences of students. Additionally, he emphasized key features that would make the app user-friendly and valuable, such as an intuitive interface and streamlined operations.

Interview #2 with Katherine ​

When interviewing Katherine, an MIT student, about their experience in academia and beyond with respect to information retention, there were many important takeaways I have noticed. Specifically, certain academic subjects are more contingent on applications of material—like mathematics and computer science—while others emphasize rote memorization more, such as those in the life sciences and foreign languages. Rine mentions that they found subjects like biology and chemistry extremely frustrating because there were not many resources available to them to help them in learning a large amount of content in a short time. An infrequent lack of avenues to apply their lecture materials towards made learning much too arduous and passive.

On the contrary, however, memorizing vocabulary when learning Japanese was much more active—facilitated with the usage of pictographs, mnemonic devices, flash cards, and games—which allowed them to engage with the material much more effectively. Rine’s ADHD makes focusing more difficult so when they employ such strategies, they are better able to retain information long-term. The span of retention varies on how often they review the material, but overall, more active participation in the learning process has proven to be a solution, like we originally proposed.

Interview #3 with Alice ​

Regarding the usefulness of creating educational songs for students, Alice agrees that songs are effective at teaching students. She provided several examples of song’s she’s used to teach students or that she used to learn subjects herself: the “ABC song,” books of the Bible to song, the months of the year, the States in the US, the elements of the periodic table. She mentioned how the church puts Psalms to tunes so people can remember them. There are songs to tell kids bible stories (“The wise man built his house upon a rock”), teach them to spell (“the B-i-b-l-e, that’s the book for me”); Educational Albums devoted to teaching kids about the bible are popular.

“When it comes to rote memorization, it’s hard to do it without a song… Put anything to song, people remember it. Like Jingles.”

Repetition makes an effective song for learning.

Interview #4 with Allison ​

The interview with Allison was full of insights and interesting takes. As a past ESL learner and current ESL teacher, Allison plays both user and stakeholder roles. Like many people Allison enjoys music, therefore she was immediately intrigued by the idea of using music for memorization/learning. Allison outlined the following points of positive impact, the app can bring to her classroom:

• Enhanced Engagement: "Students often lose interest with traditional memorization methods.” Students in ESL classrooms often have a lot of commitments outside the classroom. Turning some of the study material into songs could make learning more engaging and enjoyable. She hopes some students will be more excited to go to the class. Additionally, using the app will also be more engaging for students compared to using “boring textbooks”

• Pronunciation Practice: The app could help Allison’s students with pronunciation. Repetition through song is a fun way to practice difficult sounds and improve fluency. Some students are very self-aware of their pronunciation issues, so improving it via songs will help them out greatly.

• Memory Aid: Allison subscribes to the idea that music is a powerful mnemonic device. She said, “My students might find that lyrics stick with them longer than simple memorization using repetition."

• Cultural Connection: Incorporating popular American tunes or styles could also give students a better sense of the culture here, which is just as important as the language itself. Knowing American tunes will help them connect better with the country on a cultural level.

• Peer Learning: The app could facilitate peer learning opportunities. Students might collaborate on songs, which would help them learn from each other.

We also discussed some of the ways the app can negatively impact the learning experience of Allison’s students. Here’s the list of some effects that we want to avoid with our app:

• Over-Simplification: There's a risk of oversimplifying complex language concepts when fitting them into a song. It might not be as effective for advanced grammar structures.

• Difference in Learning Styles: While it is true that songs can help with memorization, not all students may find musical memorization effective. People have different learning styles that work for them. Some are visual learners, some have auditory learning. Those who don’t have an auditory learning style might struggle with this method.

• Potential Distraction: If the app is too entertaining, it could end up being more of a distraction than a learning tool. Students might have too much fun with songs, and get a false impression of learning. The balance between fun and education must be very clear.

Revised Impact Case ​

Addressing the issue of hard-wired memorization in academics is crucial, particularly for students facing challenges related to neurodivergence or learning disabilities. Our proposed solution is an innovative song memorization app, designed to empower students by providing them with a unique tool for long-term retention of information. Through this app, students can input academic content and effortlessly generate mnemonic songs tailored to their learning needs.

We firmly believe that our app offers a promising solution because of the wealth of evidence provided by research that supports our belief in the efficacy of utilizing songs as mnemonic devices. Furthermore, our app drawing inspiration from successful examples of mnemonic, educational songs that have garnered millions of views this one [Capitals of South America] and this one [Water Cycle Song]. These examples demonstrate that students are naturally inclined to engage with academic material when presented in an engaging and captivating manner.

To validate the effectiveness of our app, we propose conducting a comparative study. We will measure the time it takes for a student to memorize a paragraph of text in its original form versus when presented as a song generated by our app. This empirical evidence will serve as a compelling demonstration of the app's potential to enhance students' academic retention capabilities. We could also measure user engagement and improvement in retention rates among students who use the app compared to those who do not, to prove that this solution is effective in addressing the problem of limited accessibility for mnemonic devices.