Building the Future of Audio Computing

Introduction

• Lecturer's reflection on modern technology usage.
• Observations about people being glued to phones.
• Irony of criticizing excessive screen time while being on the phone.
• Query: How to create a healthier relationship with technology?

Background

• Lecturer's experience in deep tech:
  • Funded deep tech at ARPA-E
  • Worked at Google X (Google's moonshot factory)
  • Created a spin out called iyo
• Spent 10 years researching future technologies.
• Conclusion: Need for an entirely new kind of computer
  • Computer that speaks our language
  • Computer for seamless, intuitive conversations.

Introduction of Audio Computers

• 6 years of secret development.
• Characteristics of the prototype:
  • No screen
  • Entirely ear-based
  • Not just fancy earbuds but a complete computer
• Aim: Replace visual computer interactions
• Reliance on new audio-based user interface (AUI)
  • Natural language input
  • Auditory space output

Demonstration of Audio Computer

• Examples of natural language conversation:
  • Conversational interaction with the system "Q"
  • Tasks like receiving positive messages, telling jokes
• Emphasis on intuitive, less robotic interactions.

Technological Foundations

• Revolutionary aspect: leveraging natural spoken language
• Key elements:
  • Natural language
  • Superhuman processing speed
  • Access to internet and human knowledge databases
• Distinction from existing voice commands.

Potential Applications

• Examples of possible uses:
  • Conversational email briefings
  • Voice-based search
  • Personal context building through conversations
  • Less reliance on visual apps

Privacy and Wearability

• Designed for wear all day
• Enhances natural hearing
• Mixed audio-reality capabilities:
  • Enhanced ambient acoustics control
  • Ultra-high fidelity spatial sound
  • Giant audio structure for research
• Underlying research: psychoacoustics
• Reverse-engineering brain's sound positioning into software

Practical Experience Simulation

• Restaurant noise scenario
• Auditory transformations: enhacement, reduction of specific sounds, real-time translation.
• Execution of beamforming, auditory scene analysis, machine-learning denoising, AI transcription and translation, text-to-speech.

Future Possibilities

• Endless potential for app development
• Examples:
  • Personalized educational apps
  • Conversational fitness coaches
  • Relaxation soundscapes
• Developer creativity as the limit
• Aspirations to create the first audio computer and human-centric computing experience

Conclusion

• Vision: Computing interface that speaks our language intuitively
• Not monetizing attention, but enhancing life
• Invitation to move towards audio-based computing.