The challenge is to make our models good enough and easy to use so that even non-expert musicians can find them genuinely musically engaging, focusing on generating interesting sounds rather than just melodies.

We're rethinking our music sequence generation work, moving beyond simple models to care about expressive timing, dynamics, and polyphony, aiming to create high-quality music generation tools.

We haven't yet created a model that can generate an entire song because we haven't perfected the conditional aspects of the LSTM yet, which is crucial for handling long-term musical structures.

The goal is to create tools that allow composers to offload interesting aspects of their work, like expressive timing and musical texture.

Rhythmic elements are crucial for achieving the composer's intent, especially when they intricately place notes within a beat.

I find Mixolydian and Lydian scales very useful for creating interesting patterns in my music.

Playing notes on different strings presents a challenge in maintaining a consistent tone due to the different timbres of the strings.

I have models that continuously process different bodies of text and music to see what interesting outputs emerge, sometimes resulting in messy garbles and sometimes in intriguing sounds.

Adding elements like vibrato and rubato can infuse personality into the melody, making it more expressive and engaging for the listener.