Installing Ubuntu on an AMD HP Laptop

Linux for blind general discussion blinux-list at
Sun Jul 22 17:02:37 UTC 2018

Hello Harmony and all, Didier again.

My previous answer was not enough relevant, as I didn't understand
well enough your context, especially that you were using the
Gnome desktop not the console mode.

Since then I have reinstalled Ubuntu 18.04 in a virtual machine,
so I understand better your context.

First, as Speech synthesizer you have only one choice by default
which is espeak-ng. As I already wrote, don't consider dummy and
generic, they are useless in your situation.

About the voices, I have theoretically good bud practically 
bad news:

First a correction: in Ubuntu 18.04 the voices for espeak-ng are
in /usr/lib/x86_64-linux-gnu/espeak-ng-data/voices/
not in /usr/share/espeak-ng-data/voices/ as in Slint.
To know there content type:
ls -l /usr/lib/x86_64-linux-gnu/espeak-ng-data/voices/mb
ls -l /usr/lib/x86_64-linux-gnu/espeak-ng-data/voices/\!v/

This being said, I confirm that you do have all settings for
MBROLA and the KLATT variants.

But teh espak-ng binary is not installed bu default so you can't
type the commands I indicated. To install it type:
sudo apt install espeak-ng
Then all these commands work:
espeak-ng --voices
espeak-ng --voices=en
espeak-ng --voices=mb
espeak-ng --voices=mb|grep -w en
Now, theoretically to use one the variants listed in:

you should edit the file:
EspeakListVoiceVariants O
EspeakListVoiceVariants 1
Unfortunately, and this is an issue I already reported,
this puts so many lines in the combo box "Person" of the
Settings GUI of Orca that it becomes almost unusable,
especially to change the voice in this combo box.

Another possibility would be to use one of the MBROLA voices.
For this also I have also reported an issue, on which I won't
elaborate here.

Let me just say that theoretically you should install
some voices:
sudo apt install mbrola-us1
sudo apt install mbrola-us2
sudo apt install mbrola-us3
sudo apt install mbrola-us4
sudo apt install mbrola-en1

Then in /etc/speech-dispatcher/speechd.conf include these lines:
DefaultVoiceType "FEMALE1"
DefaultLanguage "en"
AddModule "espeak-ng-mbrola-generic" "sd_generic" "espeak-ng-mbrola-generic.conf"
DefaultModule espeak-ng-mbrola-generic
LanguageDefaultModule "en" 

But I am afraid that you would need a fairly recent
speech-dispatcher from git, else you will miss the file

Also, it is rather difficult to debug an issue in Ubuntu
for me as I can't find the log files for speech-dispatcher,
and I don't have the time to learn, not being an Ubuntu
user, sorry.

Didier Spaier

On 07/22/2018 01:04 AM, Linux for blind general discussion wrote:
> Hello Harmony, Didier here.
> I will answer each topic separately before quoting it.
> To know if Windows remains, type in a terminal or in a
> console:
> lsblk -l -o type,name,size,fstype,mountpoint|grep part
> If you find a partition  of type ntfs it's probably for Windows.
> If, say, it's sdb1 you can check what's in it typing as root or
> using sudo this comand:
> mount /dev/sdb1 /mnt
> ls /mnt
> Then:
> umount mnt.
> Also, if you type as root update-grub your boot loaders will be
> listed on the order they are found.
> Alternatively if the update-grub script is not shipped in your
> distribution, just type instead, either:
> grun-mkconfig -o /boot/grub/grub.cfg
> Some distributions name this command differently, then type instead:
> grub2-mkconfig -o /boot/grub/grub.cfg
> If Window is not listed, it was deleted.
>> I don't 
>> know if it actually kept Windows or if it deleted it, because when I 
>> start my computer, Linux comes right up.
> About voices; less split my answers in two parts: on the console,
> ad in a graphical environment.
> 1) On the console
> If by chance you have installed espeak-ng instead of the legacy espeak,
> you have plenty of choices especially if it was compiled with support for
> the MBROLA voice synthesizer and or the KLATT algorithm.
> To know all available voices, type one of these commands:
> espeak-ng --voices
> espeak --voices
> Both should give the same results is espeak-ng is properly installed.
> To restrict the list to English instead:
> espeak-ng --voices=en
> espeak --voices=en
> To know the available MBROLA voices, if installed:
> espeak-ng --voices=mb
> espeak --voices=mb
> You can restrict them to English like this, for instance:
> espeak-ng --voices=mb|grep -w en
> Or to get just the MBROLA voices names:
> ls -1 /usr/share/espeak-ng-data/voices/mb/
> To know the available KLATT variants:
> ls -1 /usr/share/espeak-ng-data/voices/\!v/
> To try, let's give some examples:
> You want to try the mbrola voice mb-us1, type:
> espeak-ng -vmb-us1 "I want to hear how this sentence will be spoken. Will I like the sound?" 
> For mb-	us2 instead:
> espeak-ng -vmb-us1 "I want to hear how this sentence will be spoken. Will I like the sound?"
> You want to hear English with a feminine voice (f5 KLATT variant)
> espeak-ng -ven+f5 "I want to hear how this sentence will be spoken. Will I like the sound?"
> You insist to try the feminine whispering KLATT variant:
> espeak-ng -ven+whisperf	"I want to hear how this sentence will be spoken. Will I like the 
> How to change the voice and set a default one depends on your distribution.
> The only one for which I know the answer is Slint, sorry.
> But others can step in and answer.
> 2) In a graphical environment like MATE, for instance.
> Then, orca  rely on speech-dispatcher, wichh is a speech system,
> meaning it serves to integrate (link together, insuring that they
> communicate) sevral components:
> * The screen reader (orca)
> * sound systems (e.g. pulse, libao, alsa)
> * speech synthesizers (e.g.festival, cicero, espeak-ng, pico)
> The dully syhthesizer is used for testing purposes.
> The generic synthesizer is used in conjunction with another
> component, like mbrola.
> In a graphical environment, beyond espeak-ng and mbrola you can have
> flite, festival and pico voices for free, others with a fee.
> I will let others naswer about the various possibilities
>> Also, I would like to know if there are any other speech synthesisers 
>> that work with Linux, as I don't want to just use the eSpeak default 
>> male voice. What is speech system and speech dispatcher? Also, what are 
>> the dummy and generic synthesizers?
> Sorry for this lengthy post but also thanks as it will serve as a
> basis for the documentation I have to write for Slint!
> Best regards,
> Didier
> --
> Didier Spaier
>> If you can provide me with more information that would be excellent. I 
>> am just having to get used to getting around the graphical user 
>> interface for a while until I figure out how to get both Windows and 
>> Linux running side by side without having to use a virtual machine.
>> -Harmony
>> On 07/19/2018 03:26 AM, Linux for blind general discussion wrote:
>>> In earlier versions of Ubuntu drums would play when it was ready
>>> for a login.  Maybe a way around this would be for espeak to run
>>> in command line environment to get a log in and then perhaps a
>>> user could run startx to bring up a graphical display and
>>> perhaps start orca.  You could also try orca -s after keying in
>>> alt-f2 or alternatively orca -r to replace whatever is running
>>> and start orca again from scratch.
>>> discussion wrote:
>>>> Date: Wed, 18 Jul 2018 18:11:46
>>>> From: Linux for blind general discussion <blinux-list at>
>>>> To: Linux for blind general discussion <blinux-list at>
>>>> Subject: Re: Installing Ubuntu on an AMD HP Laptop
>>>> Hello,
>>>> I tried this already but I didn't hear anything. Will it work regardless
>>>> of what sound drivers I have? I know that Narrator works just fine when
>>>> I boot from media using control Windows Enter.
>>>> -Harmony
>>>> On 18/07/2018 14:35, Linux for blind general discussion wrote:
>>>>> yes it comes with orca.  Try windowskey-s a few minutes after you start
>>>>> the computer.  You should need to do this only once.  After that, when
>>>>> you boot you'll have speech.
>>>>> On Wed, 18 Jul 2018, Linux for blind general
>>>>> discussion wrote:
>>>>>> Date: Wed, 18 Jul 2018 15:37:08
>>>>>> From: Linux for blind general discussion <blinux-list at>
>>>>>> To: "blinux-list at" <blinux-list at>
>>>>>> Subject: Installing Ubuntu on an AMD HP Laptop
>>>>>> Hello, everyone.
>>>>>> I decided to try installing a Linux distro after my Windows 10 image
>>>>>> started malfunctioning. I didn't want to have to lose all my data, so I
>>>>>> wanted to see if using Linux would allow me to access all of my files
>>>>>> and put them on an external hard drive.
>>>>>> I downloaded Ubuntu 18.04 Iso and the Rufus utility. I put the Iso image
>>>>>> on my flash drive using this utility.
>>>>>>    From past experience with an AMD HP computer, i found that pressing the
>>>>>> escape key immediately after turning on the computer, and then pressing
>>>>>> F9 would bring me to the list of boot options. From here I pressed down
>>>>>> arrow to select USB media, and pressed enter. At this point I don't know
>>>>>> what happened next because I don't know if Ubuntu 18.04 64-bit AMD comes
>>>>>> with any screen reader, or what the commands would be to launch it. I
>>>>>> used the Microsoft seeing AI app to read what was on the screen.
>>>>>> Any tips would be greatly appreciated.
>>>>>> Sincerely,
>>>>>> -Heavenly Harmony
>>>>>> _______________________________________________
>>>>>> Blinux-list mailing list
>>>>>> Blinux-list at
>>>> _______________________________________________
>>>> Blinux-list mailing list
>>>> Blinux-list at
>> _______________________________________________
>> Blinux-list mailing list
>> Blinux-list at
> _______________________________________________
> Blinux-list mailing list
> Blinux-list at

More information about the Blinux-list mailing list