I'm starting an A.I. project based on the semantics of words. I've decided to employ both TTS (Test to Speech) and SRE (Speech Recognition Engine)
For the former I've chosen to use pyttsx3 as the TTS module for Python which is the language I'm using for the A.I. project. I've come to realize that pyttsx3 actually uses espeak as the underlying TTS driver.
I should probably mention the system I"m using: I've chosen to create this project on a Raspberry Pi model 4, running Ubuntu 18.04.4 LTS operating system.
What I've done thus far:
I have successfully installed pyttsx3 and imported it into Python.
espeak was alredy installed on my Ubuntu system
The system is working in terms of speaking
I've learned how to control the rate of the speech
I've discoverd that by adding a "+f5" to the end of the language name I can make it sound somewhat female.
However, I'm not happy with the results and I would like to gain better control over the speech.
I've also downloaded MBROLA voices, but haven't yet figured out how to get espeak to use them.
What I need to learn about:
I'd like to learn how to use the MBROLA voices with espeak
I'd like to learn how to modify espeak in terms of how it actually speaks
It currently isn't prouncing words correctly and I would like to be able to modify that.
What I have just now done:
I've downloaded the zip files for the source code for espeak and espeakedit.
However, I haven't looked at the code yet and I'm not sure what I'm getting into here.
And, of course, I've posted my activities in this forum for possible feedback.
Unfortunately this forum appears to be more of a black hole than a forum.
In any case, I would love to learn of any resources that might help me with my goals.
Just for the record I'm also looking into using pocketsphinx as the SRE, Although I imagine that's an entirely different thing.
My main criteria for this TTS and SRE project
Must be Internet independent.
It's ok if it needs to be designed pretty much from the ground up.
In my A.I. project I'm going to be working with very simple words and phrases so I don't need a large vocabulary to begin with.
In fact, I would actually like to have a TTS system that I can build from the ground up, and espeak looks promising in that respect.
So if anyone reads this and can point to further information on how to modify the espeak source code and/or add MBROLA voices to the espeak system I would be very grateful if you could share that information.
And keep in mind I'm doing this on Linux Ubuntu. And I'm using pyttsx3 in Python to run espeak.
Thanks for reading.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
eSpeak project is inactive since disappearance of Jonathan Duddington.
A
ctive development and support now happens in eSpeak NG project
https://
github.com/espeak-ng/espeak-ng maintained by Reece H. Dunn.
Mailing list of espeak-ng project is https://groups.io/g/espeak-ng
This mailing list is left mostly to preserve history.
What
I need to learn about:
1.
I'd like to learn how to use the MBROLA voices with espeak
I have been watching your YouTube eSpeak tutorial videos and it's really got me interested. I'm currently working on a virtual assistant project and uses eSpeak and I would like it to teach Spanish. After watching your tutorial on the dictionary, I see that one could create a spanish dictionary for words that the eSpeak program doesn't pronounce correctly.
I was wondering if I could correspond with you to see if you have some advice on how I could use eSpeak to make a virtual assisstant spanish tutor?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello, my name's James a.k.a Robo Pi.
I'm starting an A.I. project based on the semantics of words. I've decided to employ both TTS (Test to Speech) and SRE (Speech Recognition Engine)
For the former I've chosen to use pyttsx3 as the TTS module for Python which is the language I'm using for the A.I. project. I've come to realize that pyttsx3 actually uses espeak as the underlying TTS driver.
I should probably mention the system I"m using: I've chosen to create this project on a Raspberry Pi model 4, running Ubuntu 18.04.4 LTS operating system.
What I've done thus far:
What I need to learn about:
What I have just now done:
In any case, I would love to learn of any resources that might help me with my goals.
Just for the record I'm also looking into using pocketsphinx as the SRE, Although I imagine that's an entirely different thing.
My main criteria for this TTS and SRE project
So if anyone reads this and can point to further information on how to modify the espeak source code and/or add MBROLA voices to the espeak system I would be very grateful if you could share that information.
And keep in mind I'm doing this on Linux Ubuntu. And I'm using pyttsx3 in Python to run espeak.
Thanks for reading.
eSpeak project is inactive since disappearance of Jonathan Duddington.
A
ctive development and support now happens in eSpeak NG project
https://
github.com/espeak-ng/espeak-ng maintained by Reece H. Dunn.
Mailing list of espeak-ng project is
https://groups.io/g/espeak-ng
This mailing list is left mostly to preserve history.
What
I need to learn about:
1.
I'd like to learn how to use the MBROLA voices with espeak
https://github.com/espeak-ng/espeak-ng/blob/master/docs/mbrola.md
2.
I'd like to learn how to modify espeak in terms of how it actually
speaks
https://en.wikipedia.org/wiki/ESpeak
3.
It currently isn't prouncing words correctly and I would like to be
able to modify that.
https://github.com/espeak-ng/espeak-ng/blob/master/docs/contributing.md
What
I have just now done:
1.
I've downloaded the zip files for the source code for espeak and
espeakedit.
https://github.com/valdisvi/espeak-ng-espeakedit
and
also
https://github.com/valdisvi/espeak-ng-jeditor
...
4.
Unfortunately this forum appears to be more of a black hole than a
forum.
Many
of questions in forum are about problems using Windows system. I
don't use Windows for more than 10 years
and
don't won't (and can't) help with these issues.
...
And
keep in mind I'm doing this on Linux Ubuntu.
Good!
Valdis
Hi James,
I have been watching your YouTube eSpeak tutorial videos and it's really got me interested. I'm currently working on a virtual assistant project and uses eSpeak and I would like it to teach Spanish. After watching your tutorial on the dictionary, I see that one could create a spanish dictionary for words that the eSpeak program doesn't pronounce correctly.
I was wondering if I could correspond with you to see if you have some advice on how I could use eSpeak to make a virtual assisstant spanish tutor?