VoiceToJapanese v1.1.4

Installation instructions:

Download the zip file and run the UI.exe inside.

VoiceToJapanese v1.1.4 change notes:

• Added option to change microphone mode (open mic, push to talk) and configure push to talk button
  

• writes input and translation into .txt files for displaying subtitles in OBS.

VoiceToJapanese v1.1.3

Installation instructions:

Download the zip file and run the .exe file inside.

VoiceToJapanese v1.1.3 change notes:

• updated voicevox version to include newly added voices
• fetches speakers dynamically so that all speakers/variations are avaliable

VoiceToJapanese v1.1.2

Installation instructions:

Download the zip file and run the .exe file inside.

VoiceToJapanese v1.1.2 change notes:

• runs voicevox locally by default, improving response time without needing docker etc.

VoiceToJapanese v1.1.1

Installation instructions:

Download the zip file and run the .exe file inside.

VoiceToJapanese v1.1.1 change notes:

• Moved speech to text and translation to local machine due to server cost. the program should take less than 1GB of memory and does not require GPU.
• voicevox is still hosted on azure, with the option to host locally as well.

VoiceToJapanese v1.0.0

Installation instructions:

• Download the zip file and run the .exe file inside.

VoiceToJapanese v1.0.0 change notes:

• Added the subtitler. To use, extract and run the subUI.exe file
• Added clear console button
• Added mic volume meter for troubleshooting

Subtitler demo

(please unmute the video)

subtitlerdemo2.mp4

VoiceToJapanese v0.1.1-alpha

Extract and run the UI.exe file

VoiceToJapanese v0.1.1-alpha change notes:

• The new API key is on the S0 standard plan, which allows 100 Concurrent requests by default (https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/speech-services-quotas-and-limits), this should fix the issue #4 .
• Added text input page, allowing you to synthesize longer scripts from text.
• Added the setting to use use Voice Vox from localhost. With this setting enabled, the script will try to connect to voicevox via '127.0.0.1:50021'. Running voicevox locally significantly reduces audio synthesis latency, at the expense of occupying some memory.

To run voicevox locally, follow these instructions:

Download docker:
https://docs.docker.com/get-docker/

Open command prompt and type this to start VoiceVox:

docker run --rm --gpus all -p 50021:50021 --name=voicevox  -m 4g voicevox/voicevox_engine:nvidia-ubuntu20.04-latest

VoiceToJapanese-v0.0.1alpha

Just unzip and run the .exe file inside. No additional setup required. All AI models are hosted on azure.