Daniel Tyreus

@tyreus

Co-Founder of Startups - NLP - Data Science - Product Management Engineering PhD Tinkering with and conversation-kit

San Francisco, CA
Vrijeme pridruživanja: ožujak 2009.

Tweetovi

Blokirali ste korisnika/cu @tyreus

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @tyreus

  1. Prikvačeni tweet
    1. stu 2019.

    We've been working on this one for a while and it feels great to finally have it launched! Excited to continue building.

    Poništi
  2. proslijedio/la je Tweet
    4. velj

    Attending last week convinced me that it's time for an .

    Poništi
  3. 4. velj

    Updated gif. Improved the time series analysis to better handle holes in the data. Much smoother (albeit scarier) now.

    Poništi
  4. 2. velj
    Poništi
  5. 31. sij

    Check out this great contribution to the open data model by . She applies new data to a model to estimate the potential outbreak size in before Jan 22.

    Poništi
  6. 28. sij

    “Brands will continue to struggle with discoverability on mainstream voice assistants and grapple with big tech voice platform limitations and trade-offs — data access and sharing, clunky invocations, voice synthesis inflexibility…” — 

    Poništi
  7. 26. sij

    Started a weekend hack project to look at the spread. Data from . Fork this notebook and use / contribute!

    Poništi
  8. 23. sij

    The thing I 💚 most about was getting to discuss issues with some of the greatest thinkers in the business. These are my reflections on these conversations and work with .

    Poništi
  9. 16. sij

    Enjoyed chatting about bringing content from the open web to voice assistants on the last day of

    Poništi
  10. 31. pro 2019.

    It's always nice to find that code I wrote 2 years ago still works and can bootstrap a brand new feature. This Twitter feed to voice converter has much more capability and I plan to integrate it into more topics. Thanks for not making any breaking changes!

    Poništi
  11. 16. pro 2019.
    Poništi
  12. 6. pro 2019.

    Turns out trying to do natural language summarization of calendar events is tricky. Tried to create a general algorithm that gives enough useful info without sounding wordy or unnatural. Check out the sports example in the link.

    Poništi
  13. 12. stu 2019.

    That being said, for the Voice Interoperability Initiative, having interoperability of skills across hardware devices instead of multiple wake words on the same hardware seems like a higher priority for publishers and consumers.

    Prikaži ovu nit
    Poništi
  14. 12. stu 2019.

    I was particularly interested in the "Moving up the AI Stack" strategy section on the roadmap. Making better content decisions using the customer's context would be huge for skill discovery and experience.

    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    1. stu 2019.
    Prikaži ovu nit
    Poništi
  16. proslijedio/la je Tweet
    1. stu 2019.
    Poništi
  17. 22. ruj 2019.

    Awesome meeting so many of you this weekend. Special thanks to the organizers and volunteers who help make this possible. Also shoutout to my Saturday sponsor table-sharers

    Tweet je nedostupan.
    Poništi
  18. proslijedio/la je Tweet
    Odgovor korisniku/ci

    Hey Matt! Peck plugin for 5.0 is designed to make websites responsive to voice. It includes blocks to indicate speakable content for smart speakers. Publishers can design content for voice and use WordPress as a CMS for native voice apps.

    Poništi
  19. proslijedio/la je Tweet

    How I made my ChatBot smarter by helping it understand intent, by

    Poništi
  20. 23. lip 2018.

    In full mode at our booth tracking runners. Come by and ask Google where your runner is. Or try it at home.

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·