51 lines
3.1 KiB
Plaintext
51 lines
3.1 KiB
Plaintext
|
Episode: 4172
|
||
|
|
Title: HPR4172: Re: hpr4072 Piper voice synthesis
|
||
|
|
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr4172/hpr4172.mp3
|
||
|
|
Transcribed: 2025-10-25 20:41:47
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
This is Hacker Public Radio Episode 4172 for Tuesday the 30th of July 2024.
|
||
|
|
Today's show is entitled, Read for Piper Voice Synthesis.
|
||
|
|
It is part of the series' accessibility.
|
||
|
|
It is hosted by Archers 72 and is about 5 minutes long.
|
||
|
|
It carries a clean flag.
|
||
|
|
The summary is, Archers 72 demonstrates using the OPT directory for the Piper executables.
|
||
|
|
Hello, this is your host, Archers 72 for Hacker Public Radio.
|
||
|
|
In this episode, I revisit using the Piper Voice Synthesis program, but instead of installing
|
||
|
|
it with Python PIP, I use the OPT directory.
|
||
|
|
First, I make their dashp for parent.
|
||
|
|
A directory that says it's OPT-bin Piper, and then I chone for change owner recursively to
|
||
|
|
the user, which is me on the laptop, and change the owner of OPT-bin Piper to me.
|
||
|
|
Then I go back and get clone the repo from Piper.
|
||
|
|
Before that, if I'm on a devian machine, I pseudo-apped install CMake, which is the only
|
||
|
|
dependency for making Piper.
|
||
|
|
Then I CD into Piper, type make, wait a few minutes, and the resulting files are in
|
||
|
|
the install directory.
|
||
|
|
So I copy install slash Piper asterix, which gets both Piper applications, and then
|
||
|
|
I drop it in OPT-bin Piper in a similar fashion.
|
||
|
|
I copy from the install directory, lib asterix to capture all the libraries, and copy them
|
||
|
|
to OPT-bin Piper also.
|
||
|
|
To utilize Piper, in a basic fashion, you use the ExecuteFand EXEC OPT-bin Piper slash
|
||
|
|
Piper.
|
||
|
|
You will need voices to output the text to a wave file.
|
||
|
|
You do that by pseudo-apped install git-lfs, which stands for large file system.
|
||
|
|
Then in the git directory, you git-lfs space install, and then git clone the repository
|
||
|
|
that I'll put in the show notes, and then CD into Piper voices.
|
||
|
|
From here, you can try out a voice by typing quote, this sentence is spoken first.
|
||
|
|
This sentence is synthesized while the first sentence is spoken, unquote.
|
||
|
|
Pipe that to EXEC execute slash OPT-bin slash Piper slash Piper dash dash model EN slash
|
||
|
|
EN underscore GB for Great Britain, and we're going to try a Southern English female voice,
|
||
|
|
slash low, slash EN underscore GB, Southern English female low, and it's always going to
|
||
|
|
be a dot on X, O, and NX, and then you dash dash output to voice dot wave.
|
||
|
|
Now we will demonstrate the voice by inserting the wave file from the resulting command.
|
||
|
|
This sentence is spoken first. This sentence is synthesized while the first sentence is spoken.
|
||
|
|
That's all I have for now. Feel free to record a show of your own so we can continue this
|
||
|
|
great project. This has been Archer72 for Hacker Public Radio.
|
||
|
|
You have been listening to Hacker Public Radio as Hacker Public Radio does work. Today's
|
||
|
|
show was contributed by a HBR listener like yourself. If you ever thought of recording
|
||
|
|
broadcast, you click on our contribute link to find out how easy it really is.
|
||
|
|
Hosting for HBR has been kindly provided by an onsthost.com, the Internet Archive and
|
||
|
|
R-Sync.net. On the Sadois status, today's show is released under Creative Commons,
|
||
|
|
Retribution 4.0 International License.
|