Hardware Podcasts

178: Alexa Stop

Embedded.fm · 7 Dec 2016 · 1h 09m

What this episode covers

Chris Maury, founder of Conversant Labs, discusses the design and engineering of voice-based applications, contrasting the natural language understanding of Alexa and Google Assistant with Siri's limitations. He explains the critical role of user testing in collecting diverse utterances for intent recognition and introduces TinCan.ai as a prototyping tool for non-visual interfaces. Maury also addresses the challenges of on-device speech recognition, privacy concerns, and the future of voice computing for developers and the visually impaired.

Most relevant to

early-stage companies

Topics

Mentioned

Companies
Conversant Labs, Ottomate
Products
TinCan.ai, Amazon Echo, Siri, Google Assistant, EasyVR, Grove Speech Recognizer
Tools
Pocket Sphinx

Related episodes from other shows

Other shows on the same ground, closest first.

  • IoT For All Podcast · 32m

    Implementing Voice Control in IoT | Speak To IoT's Biren Gandhi

    Biren Gandhi, Co-Founder and CEO of Speak To IoT, explains how his software platform enables voice control for IoT devices by normalizing interfaces across Alexa, Google Assistant, and Samsung Bixby. He details use cases in enterprise settings, such as Cisco and Juniper Mist integrations for population density tracking and asset location, which eliminate the need for dedicated mobile apps. Gandhi argues that voice is the 'new touch' (or 'no touch'), reducing cognitive load for users and minimizing IT overhead for manufacturers by abstracting away the complexity of multiple voice assistant APIs.

    Connected products & IoTIndustrial & consumer product design
    6 chapters

  • IoT For All Podcast · 18m

    Building Real-Time Voice AI Experiences | Agora's Patrick Ferriter | Internet of Things Podcast

    Patrick Ferriter, SVP of Product at Agora, discusses the architecture and challenges of building real-time voice AI for physical devices, such as AI companions and plush toys. He explains how Agora’s SD-RTN network and orchestration layer handle high latency, packet loss, and network degradation to maintain natural conversation flows. Ferriter details the split between edge processing (wake word, noise suppression) and cloud processing (LLMs, memory), and highlights issues like voice cloning drift and privacy concerns in multi-user environments.

    Connected products & IoTAI & machine learning
    7 chapters

  • How I Built This · 28m

    Meet The HIBT Fellows: Pierre Paul & Toby Egbuna

    Pierre Paul, founder of We Hear You, explains how his childhood experiences with exclusion inspired a sign language translator app that converts finger spelling to speech using Android tablets, aiming for full gesture recognition. Toby Egbuna, co-founder of Chezie (formerly Diversify), describes building a platform that uses curated, anonymous 'stories' and demographic 'portraits' to help job seekers from underrepresented groups find inclusive workplaces, contrasting it with generic review sites like Glassdoor. Both founders discuss their early-stage development, the ambiguity of product-market fit, and their participation in the How I Built This Fellows program.

    Starting a product companyValidation & product-market fit
    7 chapters

  • How I Built This · 45m

    AI that can be your second brain with Bethany Bongiorno and Imran Chaudhri of Humane

    Imran Chaudhri and Bethany Bongiorno, former Apple designers who worked on the iPhone, iPad, and Apple Watch, explain their decision to leave the tech giant to found Humane and build the screenless AI Pin. They discuss the engineering challenges of miniaturization, integrating multiple radios, and managing heat in a small form factor, while prioritizing privacy by avoiding always-listening wake words. The founders describe the device's functionality, including its camera, laser projector, and cloud-based 'second brain' features, and outline a business model focused on hardware sales and services rather than a traditional app store.

    Industrial & consumer product designStarting a product company
    8 chapters

  • IoT For All Podcast · 21m

    Can AI Design IoT Hardware? | Flux's Matthias Wagner | Internet of Things Podcast

    Matthias Wagner, Founder and CEO of Flux, explains how his company uses agentic AI to democratize PCB design, allowing non-engineers to create hardware via natural language prompts. He discusses the historical bottlenecks of hardware design, the necessity of iterative back-and-forth for complex devices like smartphones, and the economic advantages of designing custom boards for products like vending machines. Wagner also outlines future integration challenges, including firmware generation and enclosure design, and predicts that lowering the cost of iteration will unlock significant innovation in robotics and IoT.

    AI & machine learningPCB design & layout
    8 chapters

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by Embedded.fm. The episode itself belongs to its publisher.