Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Firefox Voice

voice.mozilla.org

Post reply on HN
Print view View original

151–160 of 166 posts

Re: Firefox Voice

#151

Post by heavyset_go »

djsumdog wrote: post #60

I've looked into open source voice assistants before. I found mycroft, Jarvis and a few others, but either got bogged down in dependencies or configuration. Many supported shipping your data to Google or Amazon if you configured it, or an open source voice recognition tool. I hate this idea that our voice has to be shipped somewhere to be processed. I remember a lot of the speech-to-text tools in the early 2000s were…

Kalliope[1] is a no code, but Python-based, voice assistant. I like it because you can switch out voice backends and it is designed to be modular with node-like constructs to build signals and responses.

I stuck it in a container and can control Home Assistant with it, run scripts, etc.

[1] https://github.com/kalliope-project/kalliope

Re: Firefox Voice

#152

Post by madacol »

No one here has mentioned this.

But I believe speech recognition will not take-off until it understands whispering speeches.

Vocal chords strain makes current solutions unsustainable for continuous usage

Re: Firefox Voice

#153

Post by michaelmrose »

sp332 wrote: post #59
michaelmrose wrote: post #50

Earlier quoted context omitted.

I'm guessing you are using the beta or nightly channel. Am I correct?

Nightly has been this way for quite a while now.

Nightly for android is in the middle of some internal changes. Full add-on compatibility with post quantum add-ons will presumably return.

Re: Firefox Voice

#154

Post by nsriv »

The Google Recorder app on Pixel phones (and I'm pretty sure general Android release) does super accurate on-device transcription, for what it's worth.

Re: Firefox Voice

#155

Post by vongomben »

synesthesiam wrote: post #118
djsumdog wrote: post #60

I've looked into open source voice assistants before. I found mycroft, Jarvis and a few others, but either got bogged down in dependencies or configuration. Many supported shipping your data to Google or Amazon if you configured it, or an open source voice recognition tool. I hate this idea that our voice has to be shipped somewhere to be processed. I remember a lot of the speech-to-text tools in the early 2000s were…

> I found mycroft, Jarvis and a few others, but either got bogged down in dependencies or configuration. More recently, there is also Rhasspy ( https://rhasspy.readthedocs.io ) and voice2json ( https://voice2json.org ). I'm the author of both, if you have questions. > Why is everything done in "the cloud." Besides being a way of collecting data and ultimately making money, it avoids some of the "bogged down in depend…

@synesthesiam your work is wonderful and I am willing to test both ASAP.

It kind of fills a gap left from snips.ai, will check and compare with what's on the market now.

Super recent (less than a year for both projects, but may be wrong) and apparently well documented. Hats off

Re: Firefox Voice

#156

Post by humaniania »

panpanna wrote: post #111
AsyncAwait wrote: post #107

I see a lot of skeptical voices here, (somewhat warranted, given it's a voice assistant technology), but the fact remains that if we want open, on-device voice recognition, we'll have to do the work and donate sample data. This extension is trying to provide some useful functionality in the hopes that Mozilla gets more data for https://commonvoice.mozilla.org I'd at least consider recording your voice, especially if…

> if we want open, on-device voice recognition, we'll have to do the work and donate sample data. Fair enough, but is any stopping Mozilla from _also_ selling your voice data to third parties including advertisers and commercial ML interests? (Asking this because Firefox has started sharing our data with leanplum)

https://support.mozilla.org/en-US/kb/how-do-you-use-leanplum...

Re: Firefox Voice

#157

Post by joecool1029 »

jasonlotito wrote: post #134
joecool1029 wrote: post #104

Earlier quoted context omitted.

Screenshot of the text I sent: https://ibb.co/X5LB4qp

I mean, that’s the last text you sent. There was clearly other text you sent that was baseless speculation. There was history there you’ve not shown. Pretending as if this is the only text that mattered is dishonest. I can see why they “banned” if you can’t even be honest to third parties.

>There was clearly other text you sent that was baseless speculation. There was history there you’ve not shown. Pretending as if this is the only text that mattered is dishonest.

Comments like this only further strengthens my point that rather than discourse this 'community' seeks to erase history and attack my credibility. Since Mozilla also censored it out and here's a screenshot of the original single text sent a day earlier: https://ibb.co/dgXBdJK

I handled all this (including attaching both screenshots) and more in my report to Mozilla. I don't believe it's dishonest to include the specific text that was censored amd banned when requested, especially when that text references the earlier 'baseless speculation' and I did not bring it up again. They asked what I was banned for, it was expressing my frustrations that caused me to send the original message in the first place.

Furthermore, its extremely concerning that the move to Matrix over IRC seems to be so that rather than just remove users they don't like, they can scrub the history of their message content and tag it with labels like 'baseless speculation' and 'conspiracy theories' to further debase the removed user.

I would include the text of the report I sent to Mozilla but it includes some personal information.

Re: Firefox Voice

#158

Post by joecool1029 »

nl8ghddbnw wrote: post #123
joecool1029 wrote: post #104

Earlier quoted context omitted.

Screenshot of the text I sent: https://ibb.co/X5LB4qp

Issue 1: "Some fields, such as "URL" and "Email Address", are privacy-sensitive and are only visible to users with minidump access."[1] - So yes, you should not send crash reports when you are dealing with sensitive data. Issue 2: That is unfortunate and should be fixed. Issue 3: The issue for Fenix is not 5 years old, it is now one year old, see [2]. Regarding your ban: - "I will not apologize for being spicy about…

Thanks for clarification on #1. It confirms what I was most concerned about and that was that my phone was leaking private info to Mozilla. On Linux I don't build the crash reporter, any crashes I used to backtrace and open with my distro first before moving to upstream. On mobile, Fennic had a specific option to turn off crash reporter, Fenix only has the two 'data sharing' sections.

I included the singular prior message posted the day before in my last comment on HN. I removed the part that was baseless in my second response which is what you saw.

Re: Firefox Voice

#159

Post by nl8ghddbnw »

joecool1029 wrote: post #158
nl8ghddbnw wrote: post #123

Earlier quoted context omitted.

Issue 1: "Some fields, such as "URL" and "Email Address", are privacy-sensitive and are only visible to users with minidump access."[1] - So yes, you should not send crash reports when you are dealing with sensitive data. Issue 2: That is unfortunate and should be fixed. Issue 3: The issue for Fenix is not 5 years old, it is now one year old, see [2]. Regarding your ban: - "I will not apologize for being spicy about…

Thanks for clarification on #1. It confirms what I was most concerned about and that was that my phone was leaking private info to Mozilla. On Linux I don't build the crash reporter, any crashes I used to backtrace and open with my distro first before moving to upstream. On mobile, Fennic had a specific option to turn off crash reporter, Fenix only has the two 'data sharing' sections. I included the singular prior me…

@ #1: To be clear, I am not a Mozilla employee. So do not take this as a confirmation, I am just quoting the documentation.

@ "Fenix only has the two 'data sharing' sections." - You can always untick the checkbox in the crash reporter, see [1]: The data choice in the settings was removed in [2] and tickboxes were added at the same time in [3]. Fenix will also remember you crash reporting choice in the following crashes, you can test this yourself using - warning, this will crash your browser - about:crashparent.

Thanks for including your previous message: I think the ban was primarily based on your first message, but that is just my personal opinion. Hopefully you get more details from your report.

Sources:

[1] https://user-images.githubusercontent.com/4037709/55157721-c...

[2] https://github.com/mozilla-mobile/fenix/issues/1264

[3] https://github.com/mozilla-mobile/fenix/issues/1265

Re: Firefox Voice

#160

Post by neop1x »

yelloworangefog wrote: post #21

I'm sorry, but cloud based speech recognition in itself would already be a red flag, even if Mozilla was doing it in-house. Outsourcing it to Google though? I feel like a company as ostensibly privacy-focused as Mozilla should really know better by now...

I feel it the same way. If they had FOSS speech to text available, then why not. But oterwise, why? For me, this voice thing is totally useless. Instead of adding it or developing new STT, they could have spend time on other innovative things. I am a big fan of Mozilla and daily user, I am still very thankful for more open and privacy friendly ecosystem it offers but implementing voice on top of Google and presenting it as a big feature looks strange to me.
Post reply on HN