OnFire Games - Love Delivery CV Voice Acting Recording

This post is a direct migration of a blog post originally written on Naver Blog on November 8, 2022.
Hello, this is Muzium.

This post introduces "Love Delivery" by OnFire Games, for which recording and post-production were carried out at Recording Studio Jium.

This is a Korean-made visual novel that emerged during a time when otome games were briefly having a moment thanks to titles like Thumbs Convenience Store and Miracle Snack Bar, and it draws from the developer's own personal experiences. It was enormously popular, selling over 10,000 copies in just about 20 days, and many streamers played through it as well. The female voice cast features well-known talent, and it received a great deal of crowdfunding support on Tumblbug — personally, I consider it the biggest hit among domestic indie games of 2021.

As of November 2022, when this post is being written, they are currently developing a new title, "Love in Login," in collaboration with Novelpia.

In October, we also recorded the demo version of "Love in Login" right here at our own studio — though details related to that are not yet something we can share!

"Love in Login" is set to launch in December!

Project Information

Client: OnFire Games

Recording Location: Recording Studio Jium

Recording Dates: December 2021 / March 2022 / May 2022

Voice Cast: Shin Onyu (Kwon Latte) / Lee Myungho (Ban Juhee) / Heo Yeeun (Shin Seungah) / Kim Heeseung (Director Park and supporting roles) / Hong Seunghyo (Supporting roles)

Mic : Neumann U87ai

Mic Pre : Millennia HV-3C

Audio Interface : Protools Carbon

The studio handled recording, direction, post-production for the voice audio, remastering of select BGM tracks, and loop processing.

Project Process

This project came about when the person in charge at OnFire Games reached out to us. They inquired about everything from recording to post-production, and after we walked them through the process and procedures, we got to work. The game launched around Christmas, and our schedule was quite hectic at the time, so we were working at a very tight pace. I recall the total recording time being around 20 hours. (That's based on records, but I think it actually took less than that, since all the voice actors made such swift progress.)

The total playtime of the game is currently around 6 hours, so you can think of it as roughly a 3x recording ratio.

Now that the project is fully wrapped up, I can say this: during an interim update, the recording was done at a different studio for reasons I won't go into — but at the request of the voice actors, they came back to our studio, Jium, to continue the work. This speaks to the fact that we always strive to provide the best possible environment for both voice actors and clients.

Aside from the publicly credited voice actors, the supporting roles were handled by voice actor Hong Seunghyo — the character count and number of roles just happened to be small enough that it wasn't as visible. If anyone reading this could add Hong Seunghyo to his Namu Wiki page, that would be great! He voiced the roles of Doctor, LD Representative, Emcee, and CEO.

Post-production was carried out to bring out the best possible voice quality for each voice actor and character.

Visual novel-style games like this one, which have a continuous story flow and significant volume variation tied to emotional expression, are particularly difficult to handle with simple Batch Processing for volume adjustment. This is because voice actors deliver everything — from soft, trailing lines like "Huh...?" to barely audible pauses conveyed through "..." — as part of their performance. When Kwon Latte curses, it comes out loud and intense; when a character cries or their emotions run high, the volume needs to follow those emotional beats accordingly.

Relying solely on Batch Processing makes the work itself much easier, but the resulting volume levels shift erratically with the story's flow and end up breaking immersion. So in the end, we approached it like listening to an audio drama without any visuals — going through everything from start to finish, carefully laying out each line in accordance with the natural flow of dialogue, and then listening through sentence by sentence with full attention.

In a visual novel, the primary media through which emotion is conveyed to the player ultimately comes down to a limited set of shared illustrations and the voice acting. That's why we're committed to delivering the highest possible quality.

Breaths within lines also need to be handled on a case-by-case basis — some are necessary for the flow, while others break immersion and need to be cleaned up individually. (Of course, the voice actors do an excellent job, but there are still one or two things that slip through here and there.)

It is through the combined efforts of so many people that a wonderful game like this was able to come into the world.