๐ฅ 6x Linkedln Top Voice | AI Research Scientist & Chief Data Scientist at IBM | Generative AI Expert | Author - Hands-on Time Series Analytics with Python | IBM Quantum ML Certified | 11+ Years in AI | MLOps | IIMA |
๐๐ฎ๐-๐ฎ๐ญ๐ต ๐๐ผ๐บ๐ฝ๐๐๐ฒ๐ฟ ๐ฉ๐ถ๐๐ถ๐ผ๐ป ๐๐ฒ๐ฎ๐ฟ๐ป๐ถ๐ป๐ด ๐๐ฎ๐ฟ๐ธ๐๐๐ก: Exploiting Knowledge Distillation for Comprehensible Audio Synthesis with GANs by Sony Computer Science Laboratories (CSL), Paris, France Follow me for a similar post:ย ย ๐ฎ๐ณ Ashish Patel Interesting Facts : ๐ธ This is a paper inย ISMIR2021 with over 1 citations. ------------------------------------------------------------------- ๐๐บ๐ฎ๐๐ถ๐ป๐ด ๐ฅ๐ฒ๐๐ฒ๐ฎ๐ฟ๐ฐ๐ต :ย https://lnkd.in/eSnUHzRk ------------------------------------------------------------------- ๐๐ ๐ฃ๐ข๐ฅ๐ง๐๐ก๐๐ ๐ธ Generative Adversarial Networks (GANs) have achieved excellent audio synthesis quality in the last years. However, making them operable with semantically meaningful controls remains an open challenge. ๐ธAn obvious approach is to control the GAN by conditioning it on metadata contained in audio datasets. Unfortunately, audio datasets often lack the desired annotations, especially in the musical domain. ๐ธA way to circumvent this lack of annotations is to generate them, for example, with an automatic audio-tagging system. The output probabilities of such systems (so-called "soft labels") carry rich information about the characteristics of the respective audios and can be used to distill the knowledge from a teacher model into a student model. ๐ธIn this work, we perform knowledge distillation from a large audio tagging system into an adversarial audio synthesizer that we call DarkGAN. ๐ธResults show that DarkGAN can synthesize musical audio with acceptable quality and exhibits moderate attribute control even with out-of-distribution input conditioning. We release the code and provide audio examples on the accompanying website. #computervisionย #artificialintelligence #data