search

mush42 / sonata-nvda

This add-on implements a speech synthesizer driver for NVDA using neural TTS models. It supports Piper
GNU General Public License v2.0
37 stars 8 forks source link

New Piper TTS Voice for NVDA #26

Closed mikebayus closed 8 months ago

mikebayus commented 8 months ago

Hi all.

I am enjoying using this add-on for the NVDA Screen reader.

I am running Windows 11 latest build, and I have just downloaded the latest voussion of the NVDA Screen Reader, V2023.3.

I have always like the voice of a Librivox reader called Amanda Friday.

Would it be possible to add her voice to the list of voices?

Here is a link where you can listen to her read.

https://librivox.org/screen-acting-by-mae-marsh/

I really really like her voice.

Thanks in advance.

mush42 commented 8 months ago

@mikebayus We're glad that you like piper-for-NVDA.

Unfortunately, we don't have the hardware resources to train a voice.

If you have the skills, you can try fine tuning an existing voice based on this using Google Colab.

Please checkout this guide for more information:

https://github.com/ZachB100/Piper-Training-Guide-with-Screen-Reader/blob/main/README.md

rmcpantoja commented 8 months ago

Yeah we can fine-tune this voice, it can be great for read another audiobooks or make audio-movies as well. Sounds great.

El mar, 31 oct 2023 a las 17:26, Musharraf @.***>) escribió:

@mikebayus https://github.com/mikebayus We're glad that you like piper-for-NVDA.

Unfortunately, we don't have the hardware resources to train a voice.

If you have the skills, you can try fine tuning an existing voice to this dataset using Google Colab.

Please checkout this guide for more information:

https://github.com/ZachB100/Piper-Training-Guide-with-Screen-Reader/blob/main/README.md

— Reply to this email directly, view it on GitHub https://github.com/mush42/piper-nvda/issues/26#issuecomment-1788120101, or unsubscribe https://github.com/notifications/unsubscribe-auth/ANATMRR4OVZVEV3VCMBLG7TYCF3JNAVCNFSM6AAAAAA6YHVWN2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBYGEZDAMJQGE . You are receiving this because you are subscribed to this thread.Message ID: @.***>

mikebayus commented 8 months ago

Thanks for the link.

It will be most helpfull.

On Tue, Oct 31, 2023 at 8:47 PM Mateo Cedillo @.***> wrote:

Yeah we can fine-tune this voice, it can be great for read another audiobooks or make audio-movies as well. Sounds great.

El mar, 31 oct 2023 a las 17:26, Musharraf @.***>) escribió:

@mikebayus https://github.com/mikebayus We're glad that you like piper-for-NVDA.

Unfortunately, we don't have the hardware resources to train a voice.

If you have the skills, you can try fine tuning an existing voice to this dataset using Google Colab.

Please checkout this guide for more information:

https://github.com/ZachB100/Piper-Training-Guide-with-Screen-Reader/blob/main/README.md

— Reply to this email directly, view it on GitHub https://github.com/mush42/piper-nvda/issues/26#issuecomment-1788120101,

or unsubscribe < https://github.com/notifications/unsubscribe-auth/ANATMRR4OVZVEV3VCMBLG7TYCF3JNAVCNFSM6AAAAAA6YHVWN2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBYGEZDAMJQGE>

. You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/mush42/piper-nvda/issues/26#issuecomment-1788229853, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKJ77NSRNQ7AQR4NHV55Z7DYCGLZDAVCNFSM6AAAAAA6YHVWN2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBYGIZDSOBVGM . You are receiving this because you were mentioned.Message ID: @.***>