Forum » Lounge » Topic
artist

AI Applied to Music

30-Aug-2024 | 10:34 PM
Any thoughts about AI and music? It seems plausible that interesting music could come from AI in the near future, yet I have my doubts... Maybe the human creativity of adjusting the control knobs would make it work? Not sure. What do you think?

https://audiocraft.metademolab.com/musicgen.html
Comments
artist
I guess we all have fears or hope for that matter. I find this letter, written by Nick Cave, very interesting and i'm sharing his thoughts (i agree 100 per cent 'a grotesque mockery of what it is to be human'), even if it's focused on songwriting and not music producing... https://www.theredhandfiles.com/chat-gpt-what-do-you-think/
3 months ago
artist
@SmokeyVW 
I listened to the songs at the link. They felt generic/academic to me. There are some hilarious AI generated songs on YouTube.
I wonder what would happen if you had AI listen to songs created by AI and songs created by humans and asked the AI to generate reviews for the tunes. Ugh... AI, AI, AI three times in one poorly constructed sentence. Musical Splenda.
@Dyl Love the Nick Cave response on the link you posted. Thanks for sharing.
Latest song: Check it Out
3 months ago
artist
@papa_g_whiz , you're right, generic/academic seems to be the aim, actually. All AI generated stuff is -by definition- made 'by the book', so it's technically perfect, pitched, synch'ed, equalized and all control knobs on the right position. And then... nothing happens. It's rather good, but empty. Quite nice, but boring. No soul, no mistake... Is it really something we need ? Did we all ask for some help in the creativity department ? No, never. Will the audience notice the difference ? I guess not. There's already tons of tunes on Spotify AI-generated, and it gets thousands of streams. It will invade our world. But we can (still) listen, support and enjoy human-made music. Even badly/quickly recorded and produced... What matters is the intent, the feeling, the genuine need to make music and art (proof is right here, on iBandstand...).
3 months ago
artist
maybe Nick Cave will write a for-real song named ‘I’ve Got the Fire of Hell in my Eyes’ just for reasons. i think that would be simultaneously hilarious and chilling.
it makes me wonder if artists (not just in music, but all creative endeavors) might start using AI generated stuff as a springboard to lead to real inspiration. sort of like using a 12-sided die to make an initial chord progression, which is then honed to make something interesting...
3 months ago
artist
heres are two answers one postive one negative.....
Postive answer...
AI can significantly enhance the music creation process in several ways, catering to both professional musicians and hobbyists. Here are some key areas where AI can assist: ### 1. **Idea Generation and Inspiration** - **Melody and Harmony Creation:** AI tools can generate melodies, chord progressions, and harmonic structures, providing a starting point for composers. These tools can analyze existing music to create new, original content that aligns with a specific style or genre. - **Lyric Writing:** AI can help generate lyrics by analyzing themes, emotions, and word patterns from existing songs or text. It can suggest words, phrases, or entire verses based on a given topic or mood. ### 2. **Music Production and Arrangement** - **Automated Arrangements:** AI can assist in arranging music by suggesting instrumentations, rhythms, and orchestrations that complement the initial ideas. It can help in layering tracks and creating complex arrangements quickly. - **Virtual Instruments and Plugins:** AI-driven virtual instruments can respond dynamically to a musician’s input, offering realistic articulations and variations that mimic live performances. AI plugins can also analyze a mix and suggest or apply effects like EQ, compression, and reverb to enhance the sound. ### 3. **Personalization and Adaptive Music** - **Custom Music Creation:** AI can create personalized music tailored to individual preferences. For example, AI can generate playlists or compositions that evolve based on the listener's mood or activity, such as workout music that adapts to your pace. - **Interactive Music Experiences:** AI can be used to create adaptive soundtracks for video games or VR experiences, where the music changes in real-time based on the player's actions or the environment. ### 4. **Collaborative Tools** - **Co-Creation Platforms:** AI can be a creative partner, offering suggestions or completing sections of a song. Musicians can collaborate with AI to explore new styles or experiment with ideas they might not have considered. - **Real-Time Feedback:** AI can provide immediate feedback on compositions, suggesting improvements in melody, harmony, rhythm, or structure. This is particularly useful for beginners learning music theory or composition. ### 5. **Music Analysis and Enhancement** - **Audio Restoration and Mastering:** AI can enhance the quality of recordings by removing noise, correcting pitch, and mastering tracks to a professional standard. It can analyze tracks to ensure they meet industry standards for various platforms. - **Music Analysis:** AI can analyze tracks for emotional content, genre classification, and even detect patterns and influences. This can be useful for musicologists or anyone studying music trends. ### 6. **Accessibility and Education** - **Music Education:** AI-driven tools can teach music theory, composition, and instrument playing more interactively. AI can adapt lessons to the learner’s pace and style, providing personalized feedback and practice sessions. - **Democratizing Music Creation:** AI tools lower the barrier to entry for music creation, enabling people without formal training to produce high-quality music. This democratization allows more people to express themselves creatively through music. ### 7. **Marketing and Distribution** - **Audience Analysis:** AI can analyze listener preferences and trends, helping artists tailor their music to their target audience. It can also assist in optimizing release strategies and marketing efforts. - **Automated Content Generation:** AI can generate promotional material, such as music videos, album art, and social media content, which aligns with the artist’s brand and style. In summary, AI's role in music creation is multifaceted, offering tools that can inspire, assist, and enhance the creative process. It empowers musicians to explore new artistic directions, streamline production, and reach wider audiences, all while expanding the boundaries of what’s possible in music.
Negative answer......
While AI has many potential benefits for music creation, it also raises several concerns and challenges that could negatively impact the music industry, artists, and the creative process. Here are some clear reasons why AI might be considered bad for music: ### 1. **Devaluation of Human Creativity** - **Loss of Artistic Authenticity:** AI-generated music could dilute the value of human creativity and originality. If AI can produce music that mimics human composition, it may lead to a flood of generic, formulaic content, reducing the appreciation for music created through human experience and emotion. - **Homogenization of Music:** AI's reliance on patterns and data can result in music that sounds similar or lacks unique, innovative elements. This could lead to a homogenization of music styles, where originality and diversity are diminished. ### 2. **Job Displacement for Musicians** - **Threat to Music Professionals:** As AI becomes more capable of creating music, it could replace human composers, songwriters, and session musicians, leading to job losses. Automated composition tools might reduce the demand for human talent in certain areas of music production. - **Erosion of Livelihoods:** Independent artists and smaller musicians might struggle to compete with AI-generated music, which can be produced quickly and cheaply. This could further concentrate the music industry around a few large corporations that can afford to deploy AI at scale. ### 3. **Ethical and Legal Issues** - **Copyright Concerns:** AI-generated music raises complex questions around ownership and copyright. If an AI generates a piece of music that closely resembles an existing work, it could lead to legal disputes over plagiarism and intellectual property rights. - **Loss of Attribution:** When AI is used to create music, it can be unclear who should be credited as the creator—the person who programmed the AI, the AI itself, or the person who curated the final product. This lack of clear attribution could undermine the recognition of human creativity. ### 4. **Cultural Impact** - **Impact on Music Culture:** Music is deeply tied to culture, reflecting human experiences, emotions, and social contexts. AI-generated music, which lacks these human elements, might not resonate in the same way with audiences or contribute meaningfully to cultural expression. - **Reduction of Musical Diversity:** AI could prioritize mainstream or popular musical patterns over more niche or culturally specific styles. This might lead to a reduction in the diversity of music available, as AI systems are often trained on data that reflects dominant trends rather than the full spectrum of musical expression. ### 5. **Dependence on Technology** - **Over-reliance on AI Tools:** Musicians might become overly dependent on AI for composing, producing, and arranging music, potentially stunting their own creative growth. This reliance could lead to a decrease in the development of traditional musical skills and techniques. - **Loss of Human Touch:** There's a risk that the emotional depth and personal touch that characterize human-made music could be lost when AI is used to automate the creative process. Music created by AI might lack the imperfections and nuances that give human music its soul and character. ### 6. **Privacy and Data Concerns** - **Data Exploitation:** AI systems often require large amounts of data to function effectively, which could include personal data about listeners' preferences and behaviors. This raises concerns about how that data is collected, used, and potentially exploited by corporations for profit. - **Surveillance and Manipulation:** AI-driven music platforms could use data to manipulate listening habits, pushing certain types of music or content for commercial gain rather than artistic merit. This could limit listeners' exposure to a wide range of music and subtly influence cultural trends. ### 7. **Environmental Impact** - **High Energy Consumption:** Training and running AI models, especially large ones used for generating music, can be energy-intensive, contributing to environmental concerns. The carbon footprint of extensive AI use in music creation could become a significant issue if not managed responsibly. ### 8. **Undermining the Value of Music** - **Commodification of Music:** If AI can generate an endless supply of music at little to no cost, it could lead to the commodification of music, where it is seen as a disposable product rather than a valuable art form. This could devalue music as a cultural asset and reduce the perceived worth of music as a profession. In summary, while AI offers new possibilities for music creation, it also poses significant risks to the value of human creativity, the livelihoods of musicians, the diversity of music, and the cultural integrity of the art form. These concerns highlight the need for careful consideration and regulation of AI's role in the music industry to ensure that it complements rather than undermines the human aspects that make music so meaningful.
Latest song: those who move
3 months ago
artist
@lapskin 
Is your response written by AI?

AI checkers think it is.
funny stuff
Latest song: Check it Out
3 months ago
artist
@lapskin  thank you for the demonstration !! The opinion of AI on AI is like AI : boring and generic !!!
3 months ago
artist
If it ain't human, I ain't interested!
Latest song: This Desert Place
3 months ago
artist
@lapskin haha. a great response. love your sense of humor!
3 months ago
artist
it was indeed ai for both ill admit but from my view point profesinaly. Sadly the whole ai thing is coming whether you like it or not me personaly untill it aactauly takes my job from me ill stick to what i do best for now  
Latest song: those who move
3 months ago
artist
Recently i have some AI guys contact me asking how i come up with certain sounds 
:D i told them am only human we go with feel ...
3 months ago
artist
@yoslounge 
"i told them am only human we go with feel ..."
---AMEN brother!
:-)
Latest song: This Desert Place
3 months ago
artist
at the moment ai cant recreate  me ive tested it each time ive broken it or it gives me some noensense i would say was made in a class room YOU CAN NOT BEAT real life world at the moment music wise art though thats a new issue 
Latest song: those who move
3 months ago
artist
person: if i have pedal point of C, what chords or melodic fragments might work well with it? 
 ChatGPT4o mini: A pedal point of C provides a stable tonal foundation that you can build various harmonies and melodies around. Here are some chord options and melodic fragments that work well: 
 Chords: 
 1. C Major (C-E-G): This is the most straightforward option, reinforcing the pedal point. 
 2. C Minor (C-E♭-G): For a contrasting, more melancholic feel. 
 3. F Major (F-A-C): The IV chord in C major, creating a nice contrast. 
 4. G Major (G-B-D): The V chord, which can lead back to C major, creating tension and resolution. 
 5. A Minor (A-C-E): The relative minor of C major, adding depth. 
 6. E Minor (E-G-B): Can provide a subtle tension before resolving back to C. 
 7. D Minor (D-F-A): The ii chord, great for adding a sense of movement. 
 8. B♭ Major (B♭-D-F): A borrowed chord that adds a rich color. 
 Melodic Fragments: 
 - C major scale: Use notes from the C major scale (C-D-E-F-G-A-B) to create simple melodies. 
 - Arpeggios: Play arpeggiated figures from the chords listed above. 
 - Stepwise motion: Moving between adjacent scale degrees can create smooth melodic lines. 
 - Repetition: Repeating a motif based on the note C can reinforce the pedal point. 
 - Chromaticism: Adding chromatic passing tones (e.g., B or B♭) can add interest.  
Tips: 
 - Experiment with rhythmic variations in your melodic lines. 
 - Use dynamics to emphasize certain notes against the pedal point. 
 - Try contrasting textures, such as a single melody line against rich chords. Feel free to mix and match these ideas to create a unique sound! 
 Person: i noticed you didn't mention A flat Major (Ab-C-Eb). i think that would work also. 
 ChatGPT4o mini: 
 Absolutely! A♭ Major can definitely work well with a C pedal point, particularly as a borrowed chord from C minor or for a modal interchange. It creates a lush, unexpected color and adds a nice contrast. 
 Using A♭ Major can evoke a more dramatic or expressive mood, especially when combined with other chords. You might also consider transitions to or from it, like moving to G minor or F minor to create even more tension and release. Great catch!
2 months ago
artist
so i guess AI can at least provide some food for thought...
2 months ago
artist
I can only comment on what I do and that is play the tenor sax and if I used AI, I am definitely cheating because it is giving me skills I do not have. Today if I use treble, echo, etc, it is is stil me playing to the level I am. I am an amateur composer and again AI would give me skills I do not have now and again I would be cheating. Bottom line I would not use AI.
2 months ago
artist
Adam Neely has some interesting thoughts on this topic   
https://youtu.be/N8NyEjB_XeA?si=LZtkjFTu6hKbTRpc  
2 months ago
artist
It all depends on what one expects from music (in general) and what one might have at stake in music, I suppose.
Audio sample loops, software synthesizers, software amps and effects are all the result of someone else's ingenuity and we use them when we make "our" music today. They are tools we use.
I've been creating my own music and playing music since the late 70's-early 80's so I looked skeptically at generative music software at first, but look... it has been only a few years since the first rough and unlistenable examples appeared and now a service like Udio can create music for you that is indistinguishable from the real thing.
It is not going to get worse from here on, it's going to get better and it's not going away. So, as much as I felt demoralized and I admit, hopeless when I first started hearing of the rapid advance of quality, I jumped in and have kept learning about generative music, art and design. AND I continue to create my own work also. So in my opinion, these generative services are still just tools (some of the best tools we've ever seen in history, but still tools). 
"A.I." has no motivation. We provide motive to these systems. We guide what they create and that improves it in the process. That will continue, no matter our personal feelings on the matter. And if your intent is to create music that you yourself are 100% responsible for, that will also continue. No one and no tool will ever take that away from any of us.
This is a song I just had made with Udio and I consider this a "collaboration" between myself and the system: https://www.udio.com/songs/4A9m8gJMS8irqE6VQc18bM
Latest song: Into Madness
1 month ago
Art Picks!

Beautiful selection of unique designs

High resolution graphics

Custom production runs available

www.artpicks.com

Search iBandstand
Welcome to iBandstand!
You'll need an iBandstand account to take advantage of many of the site's features.
Remember me