Hi, unfortunately there is no voice search build-in, and from what I can see there is no current add-on to make it work with Firefox. I guess you can rely on the platform features of Windows to use voice command, but I'm not sure how well that works.

Why is it chrome is capable and firefox isn't? (I know the answer I just want to see a moderator say it.)

Firefox doesn't have that feature. Plain enough for you?

Well that's unprofessional.

That doesn't even address my question.

What, have I personally offended you wanting to see someone official go on record with an answer?

Obviously, the browser lacks the feature.

The question is and was: Why does it lack this feature?

The further question is what can be done about it? Or am I expected to be content with a 5+ year old phone having more browsing features than a state of the art desktop browser?

Hey Innomen

I can't give you an official answer (as I don't work for Mozilla, I volunteer to help on the forums in my spare time) but I can give you my opinion.

I believe Mozilla doesn't think this is a priority feature, Different companies create different projects and invest their time into those features to separate themselves from other products. Google is known for coming out with advanced, complicated new features. And I believe that's possible due to the large number of developers they have at their disposal and their huge amount of resources to be able to test and research the voice feature until it is close to perfect.

In my opinion designing voice recognition software is one of the hardest projects I've ever seen. To create something that can work regardless of a person's accent or differing tone in their voice, to almost always correctly recognize what the person has said is amazing but also takes a huge amount of time and resources to properly test and design.

I don't think there's any shame in that Firefox can't do this yet. Other browsers don't have it, like Internet Explorer or Opera (it used to have built-in voice recognition software until April 2012 when they finally decided to discontinue the voice project. Link to story). I'm not sure why Opera dropped the Voice project. Maybe it was too costly/expensive to develop & test.

Also there's the demand for the feature. On desktop computers, I can't imagine many people using voice to search unless they have physical limitations/impairments. But on mobile phones, we know Apple's Siri and Android's Iris are very popular, high in demand features. The huge amount of work Google put into Iris helped them create speech to text search in Chrome.

Anyway, long story short. Mozilla may develop a voice search for mobile phones in its new operating system FirefoxOS or try use Android's Iris. But I don't see them investing that kind of time into voice search for the desktop browser b/c the demand isn't high enough and they probably don't have enough resources to develop one from scratch. I imagine they'll be spending their time and resources on projects that are more compatible with a desktop browser.

Also the rest of the moderators here are also volunteers. The admins though do work for Mozilla so only they could answer officially. And the-edmeister is sometimes blunt and short to the point. Don't hold it against him. :) I'm even surprised I was able to type that long of a reply. :P Anyway, hope that gave you some insight/reasons as to why Firefox doesn't have this feature.

Firstly, thank you for your serious and courteous reply. You make some good points.

Secondly, I want voice search. I would use voice search, and I'm not disabled. Arguments about why desktop users shouldn't want it or don't want it are both irrelevant to me and disproved by me.

Sidenote: Pet peeve of mine but I hate being told what I should want, as if that's an answer. Sure it gets you off the hook but it doesn't answer the question. But I see what your trying to explain, I just disregard it. I don't feel I have to take a vote before requesting a feature. But again I see what your intent was, to explain why, as I asked, not to dissuade me from asking. Just sharing the peeve since you're a moderator. :)

Thirdly, I think your argument would be valid to a degree if a certain assumption were correct. But I don't think the evidence supports it.

That assumption is that the voice processing required for the search is done client side.

Do you know for a fact that voice processing is happening within chrome (or the Google ios app) client side? If so you may skip the rest of this post and answer how you know that as I'd like to explore. :)

The fact that a ~5 year old smart phone can do it (while Dragon Speaking pretty much can't) to me shows pretty clearly that all the processing is happening server side, much like Dragon Dictation for the iPhone.

It's pretty clear to me that all chrome voice search (and Google voice search from the phone) does is stream voice to an external server and receive search results and the like as a response.

Thank you also for your note about the difference between mods and admins. Then I guess my question is directed mainly at admins or possibly mods with first hand knowledge of the correct answer.

Firefox has various or had at one time various IE spoofing technologies designed to get around Microsoft's efforts to unfairly choke out third party browsers. In my opinion getting firefox to have voice search is no more complicated than convincing the Google search page that it is the chrome browser. If the voice searching function has anything done client side I suspect that action is simple and unrelated to voice processing at all but is at most graphics and handshakes. The Google voice server protocol.

Thanks again for your serious answer. :)

Firstly, thank you for your serious and courteous reply. You make some good points.

No problem. Anytime. And sorry for the late reply.

Secondly, I want voice search. I would use voice search, and I'm not disabled. Arguments about why desktop users shouldn't want it or don't want it are both irrelevant to me and disproved by me.

Sidenote: Pet peeve of mine but I hate being told what I should want, as if that's an answer. Sure it gets you off the hook but it doesn't answer the question. But I see what your trying to explain, I just disregard it. I don't feel I have to take a vote before requesting a feature. But again I see what your intent was, to explain why, as I asked, not to dissuade me from asking. Just sharing the peeve since you're a moderator. :)

Yeah, not saying no one wants voice search. But was trying to picture how the Mozilla developers were viewing the situation. And I also don't like being told what I should want. There's a bunch of Firefox features I found useful but they are now removed (status bar [replaced by "add-on bar"], tabs on top, Properties on right-click menu, page favicon [now replaced by ugly dashed square favicon], etc). So I've definitely got a bunch of peeves with Firefox. :D

So thanks for seeing my intent. :)

Thirdly, I think your argument would be valid to a degree if a certain assumption were correct. But I don't think the evidence supports it.

That assumption is that the voice processing required for the search is done client side.

Do you know for a fact that voice processing is happening within Chrome (or the Google ios app) client side? If so you may skip the rest of this post and answer how you know that as I'd like to explore. :)

Before I was just assuming but was fairly confident it was done client side, at least for desktop users. But now I've found this Forbes' article that confirms it's done client side: Client vs. Server Architecture: Why Google Voice Search Is Also Much Faster Than Siri

The fact that a ~5 year old smart phone can do it (while Dragon Speaking pretty much can't) to me shows pretty clearly that all the processing is happening server side, much like Dragon Dictation for the iPhone.

I actually had a old phone with this feature. A Motorola flip phone v551 (pre-smartphone era) and on this phone I had discontinued service but still could use the Voice Dial feature (for saying a name or numbers). So I don't think it's fully dependent on internet service.

I think how they accomplish this and how Opera also did, is by downloading a voice library (which is probably a large size) and then doing all the voice analysis client side. It would also save them a huge amount of bandwidth if they didn't have to send a request to their servers for voice analysis everytime someone did a voice search.

See also http://windows.microsoft.com/en-US/Windows7/What-can-I-do-with-Speech-Recognition

Hey Noah_SUMO, you might want to refresh the article you linked, it now says the following:

* Correction: Although the point I was trying to make here is true in a general sense, it appears that I made an unsupported assumption about Google’s product that turns out not to be true. According to a representative from Google, in terms of the voice recognition functions of its new iOS search app, “We do very little on the device itself, and in fact the vast majority of processing goes on in the back-end. Just so happens those servers are really fast!”

Not surprising to me at all. Have you played with Dragon? It's a bulky uber slow app, and it dominates the market, also it's patented to the hilt. To think that Google had accomplished client side VR on a phone's CPU budget would have been frankly astonishing, and one would think a death stroke to Nuance/Dragon.

So I guess my original question stands.

D'oh! I have to give the author props though for at least trying to investigate how Google's Voice search works. As they've been keeping it a highly guarded secret with next to zero information on the internet about how it works. Most likely to keep Apple from improving Siri's voice search.

But there's a nitpick I have, look at the part I underlined:
According to a representative from Google, in terms of the voice recognition functions of its new iOS search app, “We do very little on the device itself, and in fact the vast majority of processing goes on in the back-end. Just so happens those servers are really fast!”

That statement applies to Google's Voice search app on iOS. Which makes sense since Apple has restrictions on what applications can do on its OS. I believe Google can't access the full capabilities of the iPhone hardware or the iOS software on the iPhone so they have to rely on server side processing there. But on Google's own Android OS, I believe they use much more client side processing since they have full access to their own hardware.

Plus here's an extra comment about it from the author under his article:
Link to author's comment

But on Google's own Android OS, I believe they use much more client side processing since they have full access to their own hardware.

I doubt it, though it is possible.

Plus here an extra comment about it from the author under his article:

A patent doesn't mean they have to use that method it just means they are the only ones that can. Given Google's experience with patent trolls tactical patenting has to be considered here. I wouldn't consider a patent evidence of anything really.

How much of the Google android app code can be seen? Would that be a way to verify client or server side or mix processing? But even if so, as you suggest, that wouldn't prove anything for ios or chrome.

I still say getting the little mic to work in firefox is little more than tricking the Google page into thinking its being read from a chrome browser. I suspect that on that note such a spoof would be more involved than simply changing an about:config setting. I'd be willing to bet there is a secret handshake somewhere.