Live data from Hacker News

Introducing Google Cloud Natural Language API, Speech API and New Data Center

cloudplatform.googleblog.com

41–50 of 51 posts

Re: Introducing Google Cloud Natural Language API, Speech API and New Data Center

#41
post #37

Earlier quoted context omitted.

>"Green" could be the color, the environmental movement, the political party, one of several people, or lots of other things It says 'train line' right in that sentence. Where is the ambiguity? even if it missed that why didn't any of the 'cta' , 'chicago' , 'train', 'run' or 'line' nudge it in the right direction. It seems to have identified 'cta' entity correctly but completely ignored that context for the next wor…

Right, this isn't ambiguous to people. But you know that CTA and Green Line are related concepts. We don't yet have a way to model all of that huge amount of common sense knowledge that people have and use to figure out what a sentence means. It's the curse of NLP, really. All the easy things are hard. (And the hard things are nigh impossible.)

As someone who's been adding schema.org to organization location pages to try and fix a possible Google Maps NLP bug[0], I can say that if Google or others ever improved their testing/analytics tools, they'd get much more adoption internet-wide on this kind of stuff. Particularly if it showed up prominently with some NLP or Google-fu inside the Developer Tools console.

I mean, if Google My Business can show me what hours and info it pulled out of my site, why can't it also suggest an editor and code snippet to use to embed that as Schema.org LD+JSON? Sorry for going off on a tangent, but it's been days since I added LD+JSON structured data and I only just yesterday learned that some parts of Google (but not the testing tool) only recognize LD+JSON inside the head of a page. There's no immediate feedback from Google whether the data I added is actually useful or not to any part of the Borg.

I think if you wanted better data, as an entity, you could easily push the engineers behind websites to give it to you. It starts with the tooling and encouragement, though.

In this specific example, I bet Google Maps via GTFS knows plenty. Now if only GTFS could be updated to use webpages and some form of Schema.org, we'd have a standard for knowledge, right? ;-) [see Footnote 3]

[0]: I'm adding the data to try and resolve a Google Maps bug where searches for "Toronto Public Library" are instantly featuring only one of the 100 branches of the library, the one closest to Toronto City Hall. Examples[1][2]. I'm now beginning to suspect NLP, since when I search for "Toronto Public Library near me" it works as intended, but when I do just "Toronto Public Library", I think it wants to find libraries closest to Toronto, and picks City Hall Branch automatically. It also is linked, strangely, to the Wikipedia entry on the Toronto Reference Library, a different branch entirely. My Schema.org is an attempt at disambiguation using parentOrganization and subOrganization references, but if the problem is the name Toronto Public Library and how Google choses to interpret that, then there's not much I can do to change it, can I?

[1]: https://www.google.ca/maps/?q=Toronto+Public+Library

[2]: https://www.google.ca/maps/?q=Toronto+Public+Library+near+No...

[3]: GTFS used to, maybe not now, but at least when I last had to read it, required mapping transit agency specifics to general Maps fields, so for example, the Heading data would lack semantic meaning but looked good in Google Maps if you put the route in it, etc.

Re: Introducing Google Cloud Natural Language API, Speech API and New Data Center

#42
post #23

Accuracy still seems to leave a fair bit to be desired. For example, when parsing "Blue, Brown, Orange, Green and Red Lines were running normally" from this news article [1], "Blue" was interpreted as the color [2], "Brown" was interpreted as the Brown Bears football team [3], "Orange" was interpreted as the Orange Line Washington (not Chicago) Metro [4], "Green" was interpreted as "environmentally friendly" [5], and…

Google Research PM here -- my team built the language understanding tech that powers the API. Thanks for checking it out! You picked a really interesting, and really hard, sentence to use to test us with. It has a couple of interesting phenomena: a reduced conjunction ("Line" goes with each color to make a name, like "Blue Line" even though "Blue" and "Line" are far apart), and high ambiguity ("Green" could be the co…

Does the technology behind this system do any classical parsing work? Or is it mostly seeing implicit structure via embeddings?

Re: Introducing Google Cloud Natural Language API, Speech API and New Data Center

#43
post #23

Earlier quoted context omitted.

Google Research PM here -- my team built the language understanding tech that powers the API. Thanks for checking it out! You picked a really interesting, and really hard, sentence to use to test us with. It has a couple of interesting phenomena: a reduced conjunction ("Line" goes with each color to make a name, like "Blue Line" even though "Blue" and "Line" are far apart), and high ambiguity ("Green" could be the co…

okay but "this is really hard" is not a good argument when ("beta" notwithstanding) this is being pitched pretty heavily by your firm with this public announcement. None of my professors or bosses have ever accepted "this is really hard" as a legitimate answer to the tasks I was given if I myself had pitched my ability to perform them. A firm like Google, with all its towering resources, is going to have a hard time…

You know nothing about NLP. Also there is a difference between good and good enough.

Re: Introducing Google Cloud Natural Language API, Speech API and New Data Center

#45
post #23

Accuracy still seems to leave a fair bit to be desired. For example, when parsing "Blue, Brown, Orange, Green and Red Lines were running normally" from this news article [1], "Blue" was interpreted as the color [2], "Brown" was interpreted as the Brown Bears football team [3], "Orange" was interpreted as the Orange Line Washington (not Chicago) Metro [4], "Green" was interpreted as "environmentally friendly" [5], and…

Google Research PM here -- my team built the language understanding tech that powers the API. Thanks for checking it out! You picked a really interesting, and really hard, sentence to use to test us with. It has a couple of interesting phenomena: a reduced conjunction ("Line" goes with each color to make a name, like "Blue Line" even though "Blue" and "Line" are far apart), and high ambiguity ("Green" could be the co…

But not even the proposed example does a great job regarding entity salience.

> Google, headquartered in Mountain View, unveiled the new Android phone at the Consumer Electronic Show. Sundar Pichai said in his keynote that users love their new Android phones.

Results in:

Google - Organization, salience: 0.27

Mountain View - Location, salience: 0.10

Sundar Pichai - Person, salience: 0.07

CES - Event, salience: 0.07

Android - Consumer Good: 0.07

So "Android" which is a central piece in the announcement ranks lower than "Mountain View", which is mostly irrelevant.

Re: Introducing Google Cloud Natural Language API, Speech API and New Data Center

#47
post #43

Earlier quoted context omitted.

okay but "this is really hard" is not a good argument when ("beta" notwithstanding) this is being pitched pretty heavily by your firm with this public announcement. None of my professors or bosses have ever accepted "this is really hard" as a legitimate answer to the tasks I was given if I myself had pitched my ability to perform them. A firm like Google, with all its towering resources, is going to have a hard time…

You know nothing about NLP. Also there is a difference between good and good enough.

I know everything there is to know about the opportunistic business practise of using the public to beta test an incomplete product, for free, and then making loads of money. Transparent.

You clearly know nothing about business ethics, nor the concept of quality.

Re: Introducing Google Cloud Natural Language API, Speech API and New Data Center

#48

Accuracy still seems to leave a fair bit to be desired. For example, when parsing "Blue, Brown, Orange, Green and Red Lines were running normally" from this news article [1], "Blue" was interpreted as the color [2], "Brown" was interpreted as the Brown Bears football team [3], "Orange" was interpreted as the Orange Line Washington (not Chicago) Metro [4], "Green" was interpreted as "environmentally friendly" [5], and…

Let's ask a rhetorical question: Who'd use this kind of information?

My answer: It's mostly search/listing apps. The biggest merit of tagging/parsing for now is that you can search the sentences who might possibly refer to some specific subway line. Toward this goal, you don't need an exact interpretation of each phrase. What you'd need is a high-level function that asks if "sentence X".refers_to("Blue Line") and the API should be designed for this kind of goals in mind.

I think the current premise that you can/should get a exact parse/tagging for a given text is kinda misleading, because an interpretation of natural language is fundamentally vague. IMO, we should go after the performance of these high-level tasks (e.g. searching or translation) rather than microscopic accuracy. It sucks that we wouldn't be able to parse the text once for all and cache the canonical representation for future use, and we'd always have to run the matching API or some sort. But I think it's kinda the nature of NLP.

Re: Introducing Google Cloud Natural Language API, Speech API and New Data Center

#49

So about $1.44 per hour, that's reasonable-ish.

We run Apache OpenNLP which gives comparable results to this service. The advantage of Google Natural Language right now is the Wikipedia link it provides to the entities it detects but I haven't seen it's results beat Open NLP

He's talking about the speech recognition. Google's speech recognition is far ahead of the competition, and also slightly cheaper (except for Baidu's which is free, but good luck getting it to work).

Re: Introducing Google Cloud Natural Language API, Speech API and New Data Center

#50
post #3

Regarding non-streaming transcription: It doesn't seem like the return includes timings, which would limit its usefulness in transcription for accessibility (where, for example, a post-processing step on our side would consume the Cloud Speech API and build SRT subtitle files for class recordings) Is there a product / service / library out there that could help align a known-good transcript with the original audio (c…

You can use Speechmatics directly
Post reply on HN