Natural Voice Fully Automated Dubbing Tool Update Ver2.1: NijiVoice Support, Batch Execution of All Steps, and More

Hello, this is Rcat.
This time, I have update information for the fully automated dubbing tool introduced below.
If you haven't already, please check this article first.

Introduction

Terms of Service

Please check the terms of service in advance when using information or works.

https://note.com/rcat999/n/nb6a601a36ef5

Regarding Comments

Please check the terms of service guidelines before posting a comment.

Update Information

NijiVoice Support

As I introduced in this article, a voice synthesis service using generative AI was announced just the other day.
It looked interesting, so I added support for this API as well, bringing the number of software options available for reading aloud to two.

Here is a test dubbing video using NijiVoice.
Since it's just after the service launch, it feels a bit subtle. Especially with decimals and such.

How to use

I have slightly modified the reading aloud part in Step 3.
I added a selection option at the very beginning, and you can use it by selecting NijiVoice here.

After that, the ID checker will switch, allowing you to confirm the character's ID. Enter the long string confirmed here as the character ID.

Also, since NijiVoice is a system charged by character count, I added an option to not use ruby.
Ruby is all in hiragana, but since the original string contains kanji, this means costs can be reduced. Since you're using AI anyway, it will read it properly, right??

Support for Batch Processing of All Steps

Until now, it was necessary to execute transcription, proofreading, reading aloud, and video creation step by step.
I will leave that as is because there are times when you want to make manual adjustments along the way, but if it's a time-consuming process, you would likely want to let it run automatically up to a certain point and leave it alone.
So, I added a feature that allows you to specify steps and execute them automatically.

How to use

Please configure all the steps you want to execute in advance.
Once all settings are complete, press the 'Execute All Steps' button added at the bottom.
You can also set how far you want to automate from the selection options above.
For example, choose Step 2 if you want to automate up to text proofreading. Choose Step 3 if you want to go up to reading aloud and get an audio file. Choose Step 4 if you want to execute until the end, then press the button.

Added a QR code for easy access from smartphones.

A QR code is now displayed at the top of this tool. You can easily access it by scanning this with your smartphone.
*Must be connected to the same local network as your computer.

Addition of an option to not use ruby characters

As mentioned at the beginning, this is a feature that reads the original text during playback, passes it to the software, and does not use the hiragana created as ruby.

This feature is a countermeasure for Niji Voice, which charges based on character count. Of course, if you feel the pronunciation in Voicevox is unsatisfactory, you are free to disable it.

Distribution Information

It is being distributed in the first article.