Custom AI voices in YMM4 (YukkuriMovieMaker4)
YukkuriMovieMaker4, YMM4 for short, is the free Windows video editor behind countless Japanese yukkuri commentary videos. It ships with AquesTalk, the classic yukkuri voice, and can use many other speech engines, from VOICEVOX and COEIROINK to SAPI5 voices and cloud services.
SuperClonica is not one of YMM4’s built-in engines, but you can still use its voices: make the lines as audio files and place them on the timeline. This guide shows how, and what to check about licenses.
What YMM4 supports itself
The YMM4 page lists built-in AquesTalk plus engines such as A.I.VOICE, AivisSpeech, CeVIO, COEIROINK, VOICEVOX, VOICEPEAK, Style-Bert-VITS2, SAPI5 voices and cloud services like Amazon Polly, Azure, Google, ElevenLabs and OpenAI. If one of those voices fits, it is the most convenient route, because subtitles and timing follow the text.
Using SuperClonica voices as audio files
Install the free engine on the same Windows PC with python -m pip install superclonica-cpu and voice a line: superclonica --text "今日はこのゲームを解説します。" --voice F2 --lang ja --output line01.wav. Save one file per line, numbered in order, and drag them onto the YMM4 timeline like any other audio. For many lines, our guide to bulk text to speech turns a spreadsheet into one file per row.
Character voices for commentary
Yukkuri videos are usually two characters talking. Give each speaker a different catalog voice, for example a calm narrator and a lively character; every catalog voice reads Japanese and 30 other languages. Any voice page plays a line of 30 to 100 characters free, so you can audition voices before you buy one.
Licenses to check
YMM4 says commercial or ad-supported videos may need a separate license, and for-profit use of AquesTalk needs its own license. SuperClonica stock voices may be used commercially; catalog character voices are fan-made, so use them for parody and fan videos and label AI voices where it matters to viewers.