course description

The current third wave of artificial intelligence is the specific achievement of modern data science. It uses big data as the material and machine learning as the processing method. It is now widely used in many fields and gradually affects all aspects of life. Taking music as an example, in addition to using classification and recommendation technology to accurately deliver products to target customer groups, it is also possible to predict the sales volume of new works before they are sold on the market. On the other hand, companies providing automatic composition services and products have sprung up at home and abroad; video games have also begun to use real-time automatic soundtracks. There is, however, still a general shortage of musical talents who can effectively communicate with data scientists and information engineers. The first half of the course introduces the brief principles of data science, and then adds the knowledge of low, middle, and high level features of music, music data representation, open music and audio data sets, etc. In the second half of the semester, through the TensorFlow Magenta artificial intelligence engine, the Google AI Hub interactive machine learning example collection, and other online free services, students can actually employ music technology such as automatic generation, assisted composition, and improvisation.

references

books:

1. A. Croll, Music Science: How Data and Digital Content Are Changing Music. Sebastopol, CA: O’Reilly Media, 2015.

2. B. D. Man, R. Stables, and J. D. Reiss, Intelligent Music Production. Abingdon: Routledge, 2019.

3. G. Mazzola, J. Park, and F. Thalmann, Musical Creativity: Strategies and Tools in Composition and Improvisation. Berlin: Springer-Verlag, 2011.

4. M. Müller, Fundamentals of Music Processing: Using Python and Jupyter Notebooks. Cham: Springer, 2021.

5. A. DuBreuil, Hands-On Music Generation with Magenta: Explore the Role of Deep Learning in Music Generation and Assisted Music Composition. Birmingham: Packt Publishing, 2020.

6. J.-P. Briot, G. Hadjeres, and F.-D. Pachet, Deep Learning Techniques for Music Generation. Cham: Springer, 2019.

courses:

1. AI For Everyone, Coursera.

2. Machine Learning for Musicians and Artists, Kadenze.

3. Music Data Science, Department of Applied Music, Tainan National University of the Arts.

feedback

anonymous feedback form

office hour

one hour every time by appointment