#rhvoice — Public Fediverse posts

La vidéo de notre dernière conférence donnée avec @dnauroy à Voxxed Days Luxembourg est désormais en ligne !

Nous y présentons un retour d’expérience sur le développement d’une synthèse vocale en luxembourgeois, adaptée aux lecteurs d’écrans.

La vidéo : https://www.youtube.com/watch?v=sEzu2l0CKr4&list=PLJgwPhYIR6h4
Les slides et le verbatim: https://gd.lu/5ZDRkh

#Accessibility #Luxembourg #OpenSource #AssistiveTech #Inclusion #A11Y #RHVoice #VoxxedDaysLU #VoxxedLU

I have just released #SpeechNote v4.9.0. This version is only for Sailfish OS. The version for Linux desktops will be released at a later date.

Release highlights:
- Support for #SFOS 5.1
- Import a note from a URL and extract only the readable text
- Switch between Notepad and Translator with a elegant tab interface (thanks to excellent Opal widget)
- New #Piper TTS and #RHVoice TTS models
- Inline timestamps in text output

Changelog: https://gitlab.com/mkiol/dsnote/-/releases/v4.9.0

Download: https://openrepos.net/content/mkiol/speech-note

I have just released #SpeechNote v4.9.0. This version is only for Sailfish OS. The version for Linux desktops will be released at a later date.

Release highlights:
- Support for #SFOS 5.1
- Import a note from a URL and extract only the readable text
- Switch between Notepad and Translator with a elegant tab interface (thanks to excellent Opal widget)
- New #Piper TTS and #RHVoice TTS models
- Inline timestamps in text output

Changelog: https://gitlab.com/mkiol/dsnote/-/releases/v4.9.0

Download: https://openrepos.net/content/mkiol/speech-note

I have just released #SpeechNote v4.9.0. This version is only for Sailfish OS. The version for Linux desktops will be released at a later date.

Release highlights:
- Support for #SFOS 5.1
- Import a note from a URL and extract only the readable text
- Switch between Notepad and Translator with a elegant tab interface (thanks to excellent Opal widget)
- New #Piper TTS and #RHVoice TTS models
- Inline timestamps in text output

Changelog: https://gitlab.com/mkiol/dsnote/-/releases/v4.9.0

Download: https://openrepos.net/content/mkiol/speech-note

I have just released #SpeechNote v4.9.0. This version is only for Sailfish OS. The version for Linux desktops will be released at a later date.

Release highlights:
- Support for #SFOS 5.1
- Import a note from a URL and extract only the readable text
- Switch between Notepad and Translator with a elegant tab interface (thanks to excellent Opal widget)
- New #Piper TTS and #RHVoice TTS models
- Inline timestamps in text output

Changelog: https://gitlab.com/mkiol/dsnote/-/releases/v4.9.0

Download: https://openrepos.net/content/mkiol/speech-note

I have just released #SpeechNote v4.9.0. This version is only for Sailfish OS. The version for Linux desktops will be released at a later date.

Release highlights:
- Support for #SFOS 5.1
- Import a note from a URL and extract only the readable text
- Switch between Notepad and Translator with a elegant tab interface (thanks to excellent Opal widget)
- New #Piper TTS and #RHVoice TTS models
- Inline timestamps in text output

Changelog: https://gitlab.com/mkiol/dsnote/-/releases/v4.9.0

Download: https://openrepos.net/content/mkiol/speech-note

Une très belle édition de Voxxed Days Luxembourg, au cours de laquelle nous avons eu l'opportunité de présenter avec @dnauroy notre projet de voix luxembourgeoises utilisables par les personnes aveugles et malvoyantes dans les lecteurs d'écrans.
Un grand merci aux organisateurs et aux participants pour les échanges très intéressants !

Pour en savoir plus sur le projet :
https://accessibilite.public.lu/fr/news/2026-03-30-screenreaderLB.html

#voxxed_lu #voxxedDaysLuxembourg #a11y #accessibility #tts #luxembourgish #luxembourg #RHvoice

@freya I know you are trying to make the most of this for you and other power users, however I just wish to add that not all the blind users are so hooked up to eloquence TTS and only a few of them can use their TTS at such high speech rates. I have understood with this initial version of Graphene Speech services we can at least independently pass the initial setup of @GrapheneOS and then we can switch to our TTS synthesizer of choice. For example I am very unlikely to stop using #RHVoice and you are very unlikely to stop using #Eloquence no matter what do the Graphene OS folks implement.
Having really good modern open-source natural sounding TTS for reading books is not a waste as you are pointing out especially if they are planning to make the other direction part of the app i.e. speech to text as well.

We have the same thing with #NVDA on windows. By default it uses natural sounding voice built into windows that has the same mistakes you are pointing out here, then it has espeak-ng included as an opensource alternative capable of running at very very high speech rates. And you are still adding eloquence into your setup eventhough you need to pass extra hoops in order to even make it run on the modern hardware and software configurations.

It's a very good thing you are advocating for all the blind users here but still if I were you I would try to calm down a bit.

@freya I know you are trying to make the most of this for you and other power users, however I just wish to add that not all the blind users are so hooked up to eloquence TTS and only a few of them can use their TTS at such high speech rates. I have understood with this initial version of Graphene Speech services we can at least independently pass the initial setup of @GrapheneOS and then we can switch to our TTS synthesizer of choice. For example I am very unlikely to stop using #RHVoice and you are very unlikely to stop using #Eloquence no matter what do the Graphene OS folks implement.
Having really good modern open-source natural sounding TTS for reading books is not a waste as you are pointing out especially if they are planning to make the other direction part of the app i.e. speech to text as well.

We have the same thing with #NVDA on windows. By default it uses natural sounding voice built into windows that has the same mistakes you are pointing out here, then it has espeak-ng included as an opensource alternative capable of running at very very high speech rates. And you are still adding eloquence into your setup eventhough you need to pass extra hoops in order to even make it run on the modern hardware and software configurations.

It's a very good thing you are advocating for all the blind users here but still if I were you I would try to calm down a bit.

@freya I know you are trying to make the most of this for you and other power users, however I just wish to add that not all the blind users are so hooked up to eloquence TTS and only a few of them can use their TTS at such high speech rates. I have understood with this initial version of Graphene Speech services we can at least independently pass the initial setup of @GrapheneOS and then we can switch to our TTS synthesizer of choice. For example I am very unlikely to stop using #RHVoice and you are very unlikely to stop using #Eloquence no matter what do the Graphene OS folks implement.
Having really good modern open-source natural sounding TTS for reading books is not a waste as you are pointing out especially if they are planning to make the other direction part of the app i.e. speech to text as well.

We have the same thing with #NVDA on windows. By default it uses natural sounding voice built into windows that has the same mistakes you are pointing out here, then it has espeak-ng included as an opensource alternative capable of running at very very high speech rates. And you are still adding eloquence into your setup eventhough you need to pass extra hoops in order to even make it run on the modern hardware and software configurations.

It's a very good thing you are advocating for all the blind users here but still if I were you I would try to calm down a bit.

@freya I know you are trying to make the most of this for you and other power users, however I just wish to add that not all the blind users are so hooked up to eloquence TTS and only a few of them can use their TTS at such high speech rates. I have understood with this initial version of Graphene Speech services we can at least independently pass the initial setup of @GrapheneOS and then we can switch to our TTS synthesizer of choice. For example I am very unlikely to stop using #RHVoice and you are very unlikely to stop using #Eloquence no matter what do the Graphene OS folks implement.
Having really good modern open-source natural sounding TTS for reading books is not a waste as you are pointing out especially if they are planning to make the other direction part of the app i.e. speech to text as well.

We have the same thing with #NVDA on windows. By default it uses natural sounding voice built into windows that has the same mistakes you are pointing out here, then it has espeak-ng included as an opensource alternative capable of running at very very high speech rates. And you are still adding eloquence into your setup eventhough you need to pass extra hoops in order to even make it run on the modern hardware and software configurations.

It's a very good thing you are advocating for all the blind users here but still if I were you I would try to calm down a bit.

@freya I know you are trying to make the most of this for you and other power users, however I just wish to add that not all the blind users are so hooked up to eloquence TTS and only a few of them can use their TTS at such high speech rates. I have understood with this initial version of Graphene Speech services we can at least independently pass the initial setup of @GrapheneOS and then we can switch to our TTS synthesizer of choice. For example I am very unlikely to stop using #RHVoice and you are very unlikely to stop using #Eloquence no matter what do the Graphene OS folks implement.
Having really good modern open-source natural sounding TTS for reading books is not a waste as you are pointing out especially if they are planning to make the other direction part of the app i.e. speech to text as well.

We have the same thing with #NVDA on windows. By default it uses natural sounding voice built into windows that has the same mistakes you are pointing out here, then it has espeak-ng included as an opensource alternative capable of running at very very high speech rates. And you are still adding eloquence into your setup eventhough you need to pass extra hoops in order to even make it run on the modern hardware and software configurations.

It's a very good thing you are advocating for all the blind users here but still if I were you I would try to calm down a bit.

Our talk about the development of a speech synthesizer for Luxembourgish has been accepted at VoxxedDays Luxembourg! 🎉
See you there!

cc @dnauroy

#a11y #rhvoice #screenReader #VoxxedDaysLU

TyfloPrzegląd Odcinek nr 321

TyfloPrzegląd Odcinek nr 321

TyfloPrzegląd Odcinek nr 321

TyfloPrzegląd Odcinek nr 321

TyfloPrzegląd Odcinek nr 321

This week in #FDroid (TWIF) is live today:

* website improvements
* more clients add banners #Flicky #NeoStore about #keepandroidopen
* #Conversations_im asks #XMPP admins to update their servers
* last #Ente Photos for x86 ABIs
* #RHVoice update
* waiting for #Orbot? Get #TorVPN
+ three new apps
& 210 updates

Get reading: https://f-droid.org/2026/02/27/twif.html

This week in #FDroid (TWIF) is live today:

* website improvements
* more clients add banners #Flicky #NeoStore about #keepandroidopen
* #Conversations_im asks #XMPP admins to update their servers
* last #Ente Photos for x86 ABIs
* #RHVoice update
* waiting for #Orbot? Get #TorVPN
+ three new apps
& 210 updates

Get reading: https://f-droid.org/2026/02/27/twif.html

This week in #FDroid (TWIF) is live today:

* website improvements
* more clients add banners #Flicky #NeoStore about #keepandroidopen
* #Conversations_im asks #XMPP admins to update their servers
* last #Ente Photos for x86 ABIs
* #RHVoice update
* waiting for #Orbot? Get #TorVPN
+ three new apps
& 210 updates

Get reading: https://f-droid.org/2026/02/27/twif.html

This week in #FDroid (TWIF) is live today:

* website improvements
* more clients add banners #Flicky #NeoStore about #keepandroidopen
* #Conversations_im asks #XMPP admins to update their servers
* last #Ente Photos for x86 ABIs
* #RHVoice update
* waiting for #Orbot? Get #TorVPN
+ three new apps
& 210 updates

Get reading: https://f-droid.org/2026/02/27/twif.html

This week in #FDroid (TWIF) is live today:

* website improvements
* more clients add banners #Flicky #NeoStore about #keepandroidopen
* #Conversations_im asks #XMPP admins to update their servers
* last #Ente Photos for x86 ABIs
* #RHVoice update
* waiting for #Orbot? Get #TorVPN
+ three new apps
& 210 updates

Get reading: https://f-droid.org/2026/02/27/twif.html

TyfloPrzegląd Odcinek nr 318

TyfloPrzegląd Odcinek nr 318

I finally got a #DeGoogled phone and it's awesome. I bought a used Pixel with #LineageOS 22.2 pre-installed. I hoped I could get voice guidance for driving directions. #OrganicMaps works great with the #RHVoice I got from #FDroid. It sounds like #FitterHappier by #Radiohead, but it beats #surveillance! I also got #Whisper+ speech recognition. It's kind of slow, but it beats surveillance! I got most of my other apps from the #AuroraStore. A few wouldn't install, but I can use the web versions.

I finally got a #DeGoogled phone and it's awesome. I bought a used Pixel with #LineageOS 22.2 pre-installed. I hoped I could get voice guidance for driving directions. #OrganicMaps works great with the #RHVoice I got from #FDroid. It sounds like #FitterHappier by #Radiohead, but it beats #surveillance! I also got #Whisper+ speech recognition. It's kind of slow, but it beats surveillance! I got most of my other apps from the #AuroraStore. A few wouldn't install, but I can use the web versions.

Working on #adacompliance by trying to integrate a #texttospeech engine into #AOSP

#espeak has been deemed too low quality, and #sherpaTTS too slow. #RHVoice silently fails on our hardware, and I'm running out of options. Does @GrapheneOS have a TTS engine that would be commercially friendly enough for us to push it via OTA update? Or would anyone recommend one?

We work with a sensitive population and lack of a screen reader can make things even more difficult for some.

Working on #adacompliance by trying to integrate a #texttospeech engine into #AOSP

#espeak has been deemed too low quality, and #sherpaTTS too slow. #RHVoice silently fails on our hardware, and I'm running out of options. Does @GrapheneOS have a TTS engine that would be commercially friendly enough for us to push it via OTA update? Or would anyone recommend one?

We work with a sensitive population and lack of a screen reader can make things even more difficult for some.

In-Process is now available, featuring all the info on CSUN ATC 2025, Thorium Reader, how the NVDA 2025.1 Update is going, Open-Source software and a new RH Voice Update! Read it all here: https://www.nvaccess.org/post/in-process-25th-march-2025/ and don't forget to subscribe via email: https://eepurl.com/iuVyjo

#NVDA #NVDAsr #ScreenReader #Accessibility #CSUNATC #CSUNATC25 #CSUN #Thorium #RHVoice #OpenSource #FOSS #FLOSS #News #Newsletter #Update

In-Process is now available, featuring all the info on CSUN ATC 2025, Thorium Reader, how the NVDA 2025.1 Update is going, Open-Source software and a new RH Voice Update! Read it all here: https://www.nvaccess.org/post/in-process-25th-march-2025/ and don't forget to subscribe via email: https://eepurl.com/iuVyjo

#NVDA #NVDAsr #ScreenReader #Accessibility #CSUNATC #CSUNATC25 #CSUN #Thorium #RHVoice #OpenSource #FOSS #FLOSS #News #Newsletter #Update

So.....

Are there any #FOSS #TTS (#TextToSpeech) apps/engines for #Android that aren't totally buggy and don't sound like TTS from the Windows 95 era?

I've got #RHVoice, but it cuts out randomly when reading entire paragraphs and sounds embarrassingly bad compared to what I had on my MacBook in 2008. :BlobCatFacepalm:

So.....

Are there any #FOSS #TTS (#TextToSpeech) apps/engines for #Android that aren't totally buggy and don't sound like TTS from the Windows 95 era?

I've got #RHVoice, but it cuts out randomly when reading entire paragraphs and sounds embarrassingly bad compared to what I had on my MacBook in 2008. :BlobCatFacepalm:

Important note for #blind users!
#CRDroid doesn't contain #gapps, there's no Talk Back available.
You have to download it manually and put it to your phone via recovery.
Also you'll need a TTS engine and this is kinda problem because I know no TTS which work out of the box without prior network access.
#Espeak needs internet to download language data on first launch;
#RHVoice needs internet for every new voice;
#Vocalizer is paid and needs the same;

The only option at least for Russian people is to download #FonTTS (modern implementation of #NewFon) and use it.
If you know totally offline TTS synthesizers for other languages, please share.

Also even if you put Talk Back and TTS via recovery they won't start via volume buttons.
You have to enable usb debugging via recovery and then, on welcome screen, via adb, enable Talk Back and set default TTS for speech output.

After that you're good to go as an usual user.

#degoogle #android #degoogled

Important note for #blind users!
#CRDroid doesn't contain #gapps, there's no Talk Back available.
You have to download it manually and put it to your phone via recovery.
Also you'll need a TTS engine and this is kinda problem because I know no TTS which work out of the box without prior network access.
#Espeak needs internet to download language data on first launch;
#RHVoice needs internet for every new voice;
#Vocalizer is paid and needs the same;

The only option at least for Russian people is to download #FonTTS (modern implementation of #NewFon) and use it.
If you know totally offline TTS synthesizers for other languages, please share.

Also even if you put Talk Back and TTS via recovery they won't start via volume buttons.
You have to enable usb debugging via recovery and then, on welcome screen, via adb, enable Talk Back and set default TTS for speech output.

After that you're good to go as an usual user.

#degoogle #android #degoogled

Hello, #fediverse!
#recommend me please some very simple and lightweight #self_hosted solution for sending email newsletters.
Recently some users of rhvoice.hu website which I maintain asked if they can receive monthly or weekly report on what's new in the project, what new versions we have or what changed.
They want very basic use case: they open a web page or send an email to a specific address and system starts to send them newsletters which we write manually or generate from our RSS feed.
If they no longer want to receive news they open a web page and unsubscribe or send an email to the same address.
Could you recommend me an open source solution for such a task?
We don't want to use proprietary services because the project is non-commercial.
Thanks in advance!

#RHVoice #RHVoice_hu #Linux #mail #selfhosted #askfedi

Hello, #fediverse!
#recommend me please some very simple and lightweight #self_hosted solution for sending email newsletters.
Recently some users of rhvoice.hu website which I maintain asked if they can receive monthly or weekly report on what's new in the project, what new versions we have or what changed.
They want very basic use case: they open a web page or send an email to a specific address and system starts to send them newsletters which we write manually or generate from our RSS feed.
If they no longer want to receive news they open a web page and unsubscribe or send an email to the same address.
Could you recommend me an open source solution for such a task?
We don't want to use proprietary services because the project is non-commercial.
Thanks in advance!

#RHVoice #RHVoice_hu #Linux #mail #selfhosted #askfedi

Сегодня выпущен хорватский язык для синтезатора речи RHVoice. Обновление для системы Андроид скоро и постепенно выйдет с обновлениями для других языков, запланированных на близкое будущее. #RHVoice #хорватский

Today i have published the croatian language for RHVoice. On android the language will roll out gradually with more updated languages i plan to make. #RHVoice #Croatian

Ура!
Подкаст, в котором мы записывались на прошлой неделе и с которым потом было так много странных историй, всё же выложен в архив вместе с кратким резюме.
Мы с @Destranis были очень рады рассказать незрячим венграм о новом синтезаторе речи, с которыми в Венгрии прям беда и грусть.

Если кому интересно, подкаст можно послушать по вот этой ссылке:
https://gephaz.hobbyradio.hu/uj-ingyenes-magyar-beszedhangok-valtak-elerhetove-a-fejlesztokkel-beszelgettunk/

Грустно, что @asael не мог участвовать в беседе, так как не знает венгерского, но мы упоминули его (правда с очень русским произношением его имени).

#RHVoice #RHVoice_hu #magyar #magyarul #podcast

Я настроил почту на #mailcow.
Это заняло всего полтора дня и то из-за того, что на сервере уже были другие проекты, а я хотел юзать не встроенный #acme от mailcow, а свой nginx на хостовой машине.

Почему-то в первый раз каждому новому провайдеру письма уходят долго, но потом видимо включается какой-то кэш, и в следующий раз всё улетает примерно так же, как у Яндекса и гугла.

Так что у rhvoice.hu теперь есть и почта.
Не знаю пока, нужно ли нам будет облако или обойдёмся без него, ибо с #nextcloud_aio я уже настрадался, а поднимать вручную тоже не очень хочется.

Но в целом моя душа довольна :)
Сейчас Ники переведёт на венгерский все страницы сайта, и можно будет идти рекламироваться в венгерские незрячие группы :)

#RHVoice #RHVoice_hu #linux #magyar #magyarul #mail

Я настроил почту на #mailcow.
Это заняло всего полтора дня и то из-за того, что на сервере уже были другие проекты, а я хотел юзать не встроенный #acme от mailcow, а свой nginx на хостовой машине.

Почему-то в первый раз каждому новому провайдеру письма уходят долго, но потом видимо включается какой-то кэш, и в следующий раз всё улетает примерно так же, как у Яндекса и гугла.

Так что у rhvoice.hu теперь есть и почта.
Не знаю пока, нужно ли нам будет облако или обойдёмся без него, ибо с #nextcloud_aio я уже настрадался, а поднимать вручную тоже не очень хочется.

Но в целом моя душа довольна :)
Сейчас Ники переведёт на венгерский все страницы сайта, и можно будет идти рекламироваться в венгерские незрячие группы :)

#RHVoice #RHVoice_hu #linux #magyar #magyarul #mail

Короче, всем, кто рекомендовал мне #Hugo, выражаю нереальную душевную благодарность!
Генератор просто огонь, есть поддержка локализации, причём разными методами, так что поддерживать сайт одновременно на английском и венгерском будет не очень сложно.
Немножко пришлось потрахаться с менюшками, но потом выяснилось, что выбранная тема кастомные меню тупо не поддерживает нормально, и нужно юзать их меню, навешивая страницам разные веса для их сортировки.
Ещё осталось разобраться с тем, как работает RSS, ибо тема вообще-то прикольная, пусть и незаслуженно позабытая многими пользователями.

После останется только поднять почту и организовать нормальное хранилище файлов, а не вот это всё (сотня файлов в одной папке, в которой и венгерский, и польский, и хорватский намешаны), и проект будет готов.

Пойду читать доку по селф-хостед почте и страдать.

#self-hosted #selfhosted #RHVoice #RHVoice_hu

Этот хрен по имени Pál Zsolt (он же Pille), реально считает, что если он нашёл в гугле датасет на Kaggle, а его друг отправил ссылку нам, то они имеют право считаться участниками и чуть ли не владельцами проекта :)
Господи, как же насрано у людей в головах.
Жаль, я не могу поделиться с вами перепиской, ибо она на венгерском, но тут такой стыд происходит, что просто жесть.
Мужику 40 с лишним лет, а он угрожает нам, что если мы не будем вести себя хорошо, как он хочет, то его команда не выпустит новые голоса для Piper TTS.
Я вам скажу там такое хреновое качество звучания, что венгерские незрячие не так много и потеряют, если он так поступит.

#pille #RHVoice #magyar #magyarul

Этот хрен по имени Pál Zsolt (он же Pille), реально считает, что если он нашёл в гугле датасет на Kaggle, а его друг отправил ссылку нам, то они имеют право считаться участниками и чуть ли не владельцами проекта :)
Господи, как же насрано у людей в головах.
Жаль, я не могу поделиться с вами перепиской, ибо она на венгерском, но тут такой стыд происходит, что просто жесть.
Мужику 40 с лишним лет, а он угрожает нам, что если мы не будем вести себя хорошо, как он хочет, то его команда не выпустит новые голоса для Piper TTS.
Я вам скажу там такое хреновое качество звучания, что венгерские незрячие не так много и потеряют, если он так поступит.

#pille #RHVoice #magyar #magyarul

4/4

Вроде бы сейчас Ники договорилась, что в подкасте будет только упомянут факт, что какой-то (без имён) венгерский фонд участвовал в записи датасета, но не более того.
Естественно никуда в лицензию к голосу на гитхабе и на других ресурсах мы этот бред писать не собираемся, ибо датасет взят легально из источника, на котором никакие фонды не упоминаются.
Но, чтобы подкаст вообще вышел в свет, пришлось пойти на вот такие уступки.

Какой же неадекватный бред, на самом деле.
И такие люди почему-то оказываются у руля и решают, чему быть, а чему не быть. Особенно актуально для множества незрячих сообществ в разных странах.

#RHVoice #magyar #magyarul #wtf