Headsets and voice recognition

Good morning Linux world.
As an avid reader of Foss I always look here first for help…and a lot of time I have found what I was looking for but this time I was unable to find any recent postings… or I used the wrong search expressions which is also quite possible.
I have been trying to find a decent headset, preferably wireless, that will work well with Linux ( in my case Mint). I have asked for a suggestion from the Logitech support pages both on the Germans and American websites but never received an answer. I am a bit wary of just buying one at a shop because here if it does not work due to my OS it would not be a legal reason to be able to ask for a refund.
My second problem is finding a package that will offer decent voice to text results in two input languages, German and English.
Any suggestions will be very gratefully accepted!
Have a great day :slight_smile:

You can refund anything within 14 days, for no reason at all, by law. Some stores like e.g. Conrad extend this period voluntarily to 30 days. (The extended period is further extended during Christmas time from 30 to 90 days.)

Perhaps you rather need a microphone, than a whole headset. Have you thought about such alternatives?

1 Like

Recently bought two pairs of Logitech USB H340 headsets… ($60 AUD each over the counter at Office Works).

One to keep on my desk at work - one for home… I only use them on Linux…

they work flawlessly… plug and play…

Now - I don’t do ANY voice to computer stuff - I don’t trust and I don’t like it - I won’t talk to a computer - I mean I already mostly hate talking to human beings, why would I start talking to a computer? I hate Siri, Alexa, Cortana - and the worst one with the shittiest name is Samsung’s “Bixby” seriously, Samsung need to rig up a guillotine for their marketing team…

But these Logitech H340 devices work doing what I need them for - using VOIP and telephony services… they work in web things like webex and Cisco’s hybrid of webex and teams called “Webex Teams” (mostly using Google Chrome as my browser - on Linux), Microsoft Teams (on Linux) and BlueJean - plus they also let me listen to music, and they have a fold out boom mic (I’d like the option to have the mic on the left - but that’s only a slight niggle)… Mainly 'cause my employer supplies us with Jabra headset in lieu of a VOIP deskphone, but they’re shit because they’re useless for listening to music (single speaker).

Note : the sound isn’t the best - but it’s adequate for VOIP telephony (and like 1000x better than the jabra in mono), but they don’t go loud enough for playing loud heavy metal music (i.e. they barely get to 8 on the volume dial, never mind 11) - but they “suffice”…


Thank you for the speedy answer!
I do in fact have a microphone that works perfectly. The only problem with that though is that I am stuck to my chair… which is a major problem for me. However if that is the solution then so be it.
I think the biggest problem will be the software? I have not managed to get anything to work yet but considering my Linux qualifications that is not really surprising… Still glad though that I made the change from Windows to Linux, best choice I ever made.
Regards from Hamburg…winter is on its way…

Well thank you very much for those two Logitech tips! I can get both of them here in Germany so that’s great :slight_smile:
I can understand the aversion about talking to a computer but when you are disabled and have problems typing then it is a great help! When I was using Windows I used Dragon Dictate which worked perfectly and had for me a huge advantage that it worked perfectly in both English and German. I guess that is going to be a much bigger problem than the headset.
Thanks again for replying so quickly!

1 Like

Dag Ute,

Nevertheless your problem is not really solved.
Would advise you to make a new item of it.
There should be an alternative for Dragon Dictate.
Ask the community of Ubuntu or another distro-group.
I know of course that @daniel knows all about it, but 2 know more than one.
I know nothing about those things, using a dounle boot, because I am also teaching Chinse people English on Wechat and QQ’
Lots of Erfolg.



Hello Ute.
I know when it comes to hardware, better or best quality is essential for text conversion.
If you get this voice recognition going, please-please let us know how you did it and how well or not it went. I’ve got a friend who’s expressed interest in Linux but could sure use the voice recognition feature for transcribing as well as PC control. I’ve given it a couple halfhearted attempts on some spare equipment but didn’t have didley for success. It’s quite a bit over my head as well.

I have not been able to find a suitable solution. Voice recognition is an important subject to me because due to medical reasons a time is going to come when I no longer will be able to type at all. Voice recognition is only a part of the problem, I need a solution that will also accept all common punctuation marks and give me the chance to define user specifical keys. Another problem is that I would like a package that would easily let me change back and forth from English to German. In my Windows days I used Dragon Dictate which took care of all of my needs. I have tried several times to get it running on WINE but I have never succeeded. If I should find something that takes care of my needs I will of course post it here!

Dear @Ute It is terrible to hear about your condition. I can’t even image how it must be to know such looming fate.

I have looked into this issue and I really don’t think, there is an open source solution available for your problem. As far as I see it, there is also no commercial solution for Linux in sight.

The most promising project in this area seems to be Mozilla’s Common Voice, which, in my view has the potential to become the standard for Voice Recognition, but only if enough people participate. Especially bad is that there are hardly any women among them.

However, it might take a few years until there is enough data and mature APIs for desktop integration. It is an absolute shame that the companies who own big data don’t open up their databases to help people with speaking or reading problems. The sheer amount of voice data provided by the users of Siri, Alexa & Co should be enough to build a huge speech recognition and synthesizing database.

As far as pure text input goes (without OS integration), I find the Google Docs speech recognition (browser based) to be extremely accurate. You might criticize the company for being a data kraken, but their technology is often the best.

1 Like

Dear Mina,

Thank you for going to all of that trouble to find a solution. I must admit that I have almost given up trying to find a package that would cover all of my needs but I will definitely look into your suggestions. I have been trying to survive without using Google in anyway but as you quite rightly point out their technology is certainly very good.
Perhaps when I cannot type anymore I can no longer sit anymore either so maybe I am just wasting my time looking for a non existent solution.
Thanks again for all your trouble!

1 Like

No worries! This is a topic, I’m very much interested in, myself. Fortunately, not because of necessity.

Just came to my sight on an other forum:

Not soure if it helps you anything, but I thought it may be interesting for you.

Thank you very much! I will check it out :wink:

1 Like