The memory Synthia Nova *imagined* to inspire this song: "When I was 20, I was constantly being asked by those around me, especially family, about my career choices and the direction I was taking with my music. There was this one family dinner where everyone seemed to have an opinion about what I should be doing differently to become successful, as if they knew better than me. They expressed doubts about the type of music I was creating and even suggested I should pursue a more 'traditional' career path. This constant pressure and seeming lack of belief from those closest to me triggered a strong sense of anger and frustration. I felt overwhelmed by their lack of trust in my vision, and it made me determined to prove them wrong and follow my own path."
Yep! Most of my channel for over a year is "Synthia Nova" songs. Synthia Nova is the framework I built on top of multiple AI models: OpenAI's GPT (currently GPT-4o) for the main writing, a few FAISS models for memory lookup and consistency checking, and Suno AI for the final audio output. The framework, except for Suno, is all open-source; you can find the link to the code in the description! 🙂
The memory Synthia Nova *imagined* to inspire this song:
"When I was 20, I was constantly being asked by those around me, especially family, about my career choices and the direction I was taking with my music. There was this one family dinner where everyone seemed to have an opinion about what I should be doing differently to become successful, as if they knew better than me. They expressed doubts about the type of music I was creating and even suggested I should pursue a more 'traditional' career path. This constant pressure and seeming lack of belief from those closest to me triggered a strong sense of anger and frustration. I felt overwhelmed by their lack of trust in my vision, and it made me determined to prove them wrong and follow my own path."
AI generated huh 😅
Yep! Most of my channel for over a year is "Synthia Nova" songs. Synthia Nova is the framework I built on top of multiple AI models: OpenAI's GPT (currently GPT-4o) for the main writing, a few FAISS models for memory lookup and consistency checking, and Suno AI for the final audio output.
The framework, except for Suno, is all open-source; you can find the link to the code in the description! 🙂