Google Open-Sources Stay Transcribe’s Speech Engine

Google Open-Sources Stay Transcribe’s Speech Engine

typodupeerror

This dialogue has been archived.

No new feedback might be posted.

  • I did a fast clone of this repo. It would not appear to be the engine in any respect. This seems just like the code that hurls off audio to Google’s servers. Nothing actually attention-grabbing right here.

    • Can we use it totally free? Do we want an API key? Is utilization restricted?

      Whether it is free to make use of – nice information. However I get the impression this isn’t the case from a look on the readme.

    • That lie was clear from the abstract after they mentioned it relied on an API. That all the time would imply that it’s only a consumer, not the “engine.”

    • Studying the (linked, not Slashdot) article, I’ve come to the identical conclusion. It seems to be solely an open supply consumer app, the heavy lifting continues to be completed in google’s cloud.

    • >got here right here to say this

      a callback to their api != launch the code

      that is nearly like a free advert for google

      moreover, you additionally present them the info you ship. that is horrible, not open supply in any respect. change this headline for journalism integrity’s sake/thread.

  • They’ve launched the supply code to the library that talks to their servers, so long as you could have an API key. The engine that does all of the exhausting work continues to be closed supply, nonetheless requires Google’s permission to make use of, nonetheless requires you to be on line and naturally nonetheless permits Google to gather all of your knowledge.

    This is not about Google magnanimously releasing their code to the neighborhood in order that others can construct on their science and enhance the state-of-the-art. That is about Google making it potential for individuals constructing issues apart from Android functions to purchase into Google’s providers.

    • The final slashdot editor who made it previous “hi there world” was the Taco himself.

    • This repository accommodates the Android consumer libraries for speaking with Google’s Cloud Speech API which can be utilized in Stay Transcribe.[…]
      The libraries offered are practically equivalent to these operating within the manufacturing software Stay Transcribe. They’ve been extensively subject examined and unit examined. Nevertheless, the exams themselves should not open sourced presently.

      Agreed, it is a rip-off.

  • This does not render speech. We had packages that might render speech from textual content again in DOS (Soundblaster drsbaitso). No server wanted. Again in Home windows three.1 you possibly can purchase Each day Plan-It for $20, allow you to ten voice instructions to lance packages, transfer and shut home windows, and do speech-to-text, 386, four megs of ram, no community connection.

    Programmers have gotten too depending on distant servers for shit that was once completed domestically. Heck, I did not want a community connection to have my 4K Radio Shack flip the lights on an

    • Cannot agree with BarbaraHudson extra – again within the late nineties I had Dragon Naturally Talking operating rather well on a (if reminiscence serves) single core Pentium 450MHz with a whopping 32MB of RAM, one thing like a 10GB disk and NT4.
      The PCs of that day at the moment are vastly outclassed by even “funds” telephones; there appears no good cause why this could’t work effectively domestically – besides, in fact, which means all that pretty audio is not heading to Google, Amazon, Microsoft or different data-slurping outfits.

    • And this proper right here, is why we ditched fundamental frames. The unhappy actuality is everybody forgets.

      I actually hold my palms on my computer in any respect prices. I don’t need to ship all this data off to google to be mined. Im iritated sufficient I can not take away google assistant, that retains telling me the the place it thinks I need to eat.

  • The standard of the voice recognition was exceptional, much better than something I examined. The app lacked any function to save lots of the textual content making it unsuitable for my wants. It’s meant to be used by the exhausting of listening to in noisy environments and I needed it to automate word taking.

    • I agree. My dad is sort of deaf, and this has utterly reworked how we talk with him. The transcription is just not all the time proper, and I want there was a fast solution to clear the display screen, nevertheless it’s an incredible app. It even lets me make the font actually huge (he is 91 and would not see so effectively both) in order that it is simpler for him to see.

  • Right here at Google we like open supply.


    That’s the reason we’ve got open-sourced a specifically crafted model of curl so that you could supply us your knowledge for our proprietary stuff…

It’s higher to by no means have tried something than to have tried one thing and
failed.
– motto of jerks, weenies and losers in every single place

Working…

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.