Watson now brings cognitive speech capabilities to developers
developer.ibm.com
developer.ibm.com
We tried using the "tradeoff Analytics" service for the project - and I must say , the tools and help available around it, the API and its documentation is pretty bad , convoluted and unusable. This is true too other services available through watson too.
We looked into IoT ( Internet of things ) as well . And once again , ran into a ton of dead ends without being able to proceed. The API documentation and examples just suck . If you are used to playing around with well documented APIs / Tools / Languages - this is going to be frustrating.
If the OP is the person who actually wrote the article , please please please go back to Watson Dev Cloud or BlueMix and try it out your services and APIs an a consumer.
In my last conversation over the phone, trying to set up the tools to work with Bluemix, several "Getting Started" pages for important steps lead to 404 errors.
For a Watson-specific example, at least a few months ago it was the case that putting together a full-featured client implementation for Q&A required poking around several obscure webpages and then plenty of runtime experimentation on top.
1) Decouple your examples from Bluemix. It's needlessly complicating your efforts to get people using the APIs which is why people are showing up, not because they want a PAAS. At least not right off the bat. Less moving parts = better.
2) Create GOOD examples/API libraries for all the languages you're creating these examples for. The ruby on rails example is lacking. It doesn't demonstrate how to call the APIs at all. I'm using the REST API docs and the only headway I've made there is that I'm able to authenticate and create a session. Trying to upload audio using your documented approach (passing the session id as a path param) returns a 401. Passing the u/p via BasicAuth in the same request doesn't work either. As of right now I have no idea how to get an audio post to work and very little in the way of debugging information/docs.
I did get things working and am working on a ruby gem that I'm hoping to be able to release for general consumption but in the mean time for folks using ruby here's a snippet that will get the simple case working once you've setup a bluemix app and bound the speech to text service to it:
Are there plans to open up parts of the older voice technology and contribute it e.g. to CMU Sphinx?
Speech to Text: https://speech-to-text-demo.mybluemix.net
Text to Speech: http://text-to-speech-demo.mybluemix.net
> it might just be good enough that I can <project> into it on that one.
Funny. It doesn't stress <project> as a verb, but as a noun, making the whole phrase mean something else.
Alex has a list of words like that too: live (to live) and live (live concert), progress (also verb and substantive), record, suspect and a bunch of other words that have multiple pronunciations based on the surrounding words (they are called homophones).
They should add homophone disambiguation - probably solvable with a classifier based on features extracted from surrounding words and POS tagging.
http://www.msnbc.com/msnbc/how-supercomputer-sees-the-state-...
What the folks @ MSNBC did was to pass the last 10 SOTU speeches to Watson and collate the results over a graph.
But, Here is why I have trouble believing Watson's perception. Try passing the following input to Watson - (or any other gibberish)
"jkldsjglkfdsjgdfls kg;jsf g dsfg fdg jsdfjgdfskg dfsgj df gfgdflg;dfg g;fkgljsdfgk;gjldfg dgkjldfgdhfgkjdfhjg fkldskf;ksdlf;ksdlfks jkdhfkhsdjkfhksdhj ljfsdjfhdsjkfjskdhfkjsdhf sdfkls;dkfl; dkfl;sd;fsk roweruoweuroiweuroiwe uweoruweoruweo ruweuro kjgsfgjkldfsjgs klfgjfdsl gjfdlkg fd jkldsjglkfdsjgdfls kg;jsf g dsfg fdg jsdfjgdfskg dfsgj df gfgdflg;dfg g;fkgljsdfg kjgsfgjkldfsjgs klfgjfdslgjfdlkg fd jkldsjglkfdsjgdfls kg;jsf g dsfg fdg jsdfjgdfskg dfsgj df gfgdflg;dfg g;fkgljsdfg kjgsfgjkldfsjgs klfgjfdsl gjfdlkg fd jkldsjg lkfdsjgdfls kg;jsf g dsfg fdg jsdfjg dfskg dfsgj df gfgd flg;dfg g;fkglj sdfg kjgs fgjk ldfsj gs klfgjfds lgjfdlkg fd jkldsjg lkfdsjgdfls kg;jsf g dsfg fdg jsdfjg dfskg fsgj df gf gdflg;dfg g;fkglj sdfg kjgsfgjkldfsjgs klfgjfd slgjfdlkg fd jkldsjglk fdsjgdfls kg;jsf g dsfg fdg jsdfjgdf skg fsgj df gfgdflg;dfg g;fkgl jsdfg kjgsfgjk ldfsjgs klfgjfdslgjfdlkg fd jkldsjg lkfdsj dfls kg;jsf g dsfg fdg jsdfjgdfskg dfsgj f gfgd flg;dfg g;fkgljsdfg kjgsfgjkldfsjgs
fkldskf;ksdlf;ksdlfks jkdhfkhsdjkfhksdhj ljfsdjfhdsjkfjskdhfkjsdhf sdfkls;dkfl; dkfl;sd;fsk roweruoweuroiweuroiwe uweoruweoruweo ruweuro kjgsfgjkldfsjgs klfgjfdsl gjfdlkg fd jkldsjglkfdsjgdfls kg;jsf g dsfg fdg jsdfjgdfskg dfsgj df gfgdflg;dfg g;fkgljsdfg kjgsfgjkldfsjgs klfgjfdslgjfdlkg fd jkldsjglkfdsjgdfls kg;jsf g dsfg fdg jsdfjgdfskg dfsgj df gfgdflg;dfg g;fkgljsdfg kjgsfgjkldfsjgs klfgjfdsl gjfdlkg fd jkldsjg lkfdsjgdfls kg;jsf g dsfg fdg jsdfjg dfskg dfsgj df gfgd flg;dfg g;fkglj sdfg kjgs fgjk ldfsj gs klfgjfds lgjfdlkg fd jkldsjg lkfdsjgdfls kg;jsf g dsfg fdg jsdfjg dfskg fsgj df gf gdflg;dfg g;fkglj sdfg kjgsfgjkldfsjgs klfgjfd slgjfdlkg fd jkldsjglk fdsjgdfls kg;jsf g dsfg fdg jsdfjgdf skg fsgj df gfgdflg;dfg g;fkgl jsdfg kjgsfgjk ....
=====
And Watson rates it as the following.
Big 5
Openness100% Adventurousness100% Artistic interests2% Emotionality1% Imagination100% Intellect100% Authority-challenging100% Conscientiousness93% Achievement striving94% Cautiousness57% Dutifulness1% Orderliness1% Self-discipline81% Self-efficacy3% Extraversion1% Activity level1% Assertiveness1% Cheerfulness1% Excitement-seeking2% Outgoing1% Gregariousness1% Agreeableness1% Altruism1% Cooperation1% Modesty1% Uncompromising1% Sympathy1% Trust1% Emotional range11% Fiery1% Prone to worry10% Melancholy34% Immoderation24% Self-consciousness6% Susceptible to stress9%
Needs
Challenge61% Closeness84% Curiosity51% Excitement66% Harmony65% Ideal54% Liberty75% Love23% Practicality86% Self-expression25% Stability60% Structure57%
Values
Conservation78% Openness to change5% Hedonism15% Self-enhancement76% Self-transcendence11%
========
Its plain gibberish , and you still get some results. I tried passing other text transliterated to English and Watson still gives results like this. I would expect it to atleast call it out as gibberish-text.
We have an update coming for User Modeling (to be announced soon). After that update, such a gibberish post will return an error.
User Modeling is based on word counting. Users should ensure that their input is actually from a human and intelligible. The service looks for certain words in the input, and will reject input that doesn't have enough of those words for the service to estimate characteristics. In the upcoming release, the documentation will explain how this works and what the relevant words are.
Also, we will provide a measurement of how accurate our results are based on the number of words that are in the input. This should allow users to understand the reliability of the results in the context of their application (e.g. a casual movie recommender app might be ok with very low confidence, while an application that makes more critical recommendations might require higher confidence).