Once I had locked in my flow, my script, and my puzzles, I was finally ready…to record.

Oh boy.

Oh boy oh boy oh boy. This ended up being…quite the process. Get ready for mostly GIFs…because, WordPress won’t let me upload videos of the tape recordings.

Recording was the thing that was on my mind all throughout the process but that I had no clue how it would truly go until I actually did it.

The very first decision I had to make was whether I would attempt to record directly onto the tape recorders I had purchased or record digitally and then somehow record those digital files to tapes after the fact.

A couple of tests with the tape recorder gave me the answer pretty quickly. First, recording through the tape recorder would require me to coordinate with another person to physically meet me as, well, there are two characters in this story. It would also just be too difficult to get a perfect recording in one take with some of the sound effects I needed. Added to all of that is that when I listened back, there was quite a bit of background static that was just not enjoyable to listen to.

I opened up my Spotify playlist and played a random song next to the tape recorder from my Google nest speaker. It seemed to sound pretty decent so I figured…hey, why not record the digital files to tape the same way I made mix tapes in the 90’s? Literally hold the recorder up to a speaker. I’d done a couple of tests. Should be fine, right?

That decision would almost end up costing me my sanity in the project. More on that in a bit…

The next choice I had to make was to find a voice actor to be Anna. I had chosen the role of Becca for myself mostly because it had more recording and if something went wrong with a particular puzzle, it was easier for me to re-record. For the part of Anna, I immediately went to Stephanie Bergeron, my very talented voice actor/actor/dancer/gamer friend who I knew would be great for the role as well as knowledgeable about audio recording in general.

Thankfully for me, Steph was interested. And of course I was going to pay her. Pay your creatives, everyone! I enjoy working with Steph because her style, much like my own, is collaborative and I welcomed her input on some of the lines and recording advice. For recording, I chose to set up a temporary account with Zencastr, a cloud recording software that was great for me in the past for podcasting. The benefit with Zencastr is that it records everyone in the room at the same time, so syncing isn’t nearly as much of an issue. You can also download all of the separate tracks afterward in both WAV and MP3 format! Apart from a couple of technical hiccups, it worked great. Of course, we both recorded on Audacitiy for safety. Because we are paranoid.

Both of us were on vastly different mics, Steph’s being a more professional grade mic for her voice acting and mine being a cheap condenser mic I used for podcasting and streaming (I really need a new mic). But we both figured that since the recordings were just going to a tape afterward, quality and levels weren’t as much of an issue. We ended up doing two takes of every tape, occasionally taking moments if we screwed up to pause and start again. Yup, digital recording was the right way to go.

It also felt good to act again. I think the one thing we struggled with sounding young enough and portraying the age jumping from age 14 to 18 through the course of the game. I think we did pretty well all things considered, and Steph is great to play off of. One thing I would want to work on in the future is my directing abilities which are…limited. Steph was able to do a lot with the little I was able to give her, though.

The other thing I had to record was a playlist of 90’s songs for each tape. This took a shockingly long amount of time…at least, shocking for me, I don’t make a lot of playlists. I got about an hour of songs for each tape and made sure that they were released at the time the story took place. Then I realized I didn’t want to have songs on both sides of the tape to make it clear where the puzzles were. So I had to edit those playlists down to 30 minutes. Some of the eliminations were easier than others once I remembered how problematic some of the pre-21st century lyrics were. But I managed to get the lists winnowed down!

I decided I would make a recording of all of the music together since I did have some voice recording to layer over top of one song per tape and it was just easier to have everything together. To record, I used OBS, a streaming software, to record the Spotify playlist playing right off my computer. The challenge with this was making sure some windows notification didn’t pop in and make it into the recording. Which…it did on occasion.

After the recording was done, it was time to edit. First I edited down the scripted portions. I use DaVinci Resolve, a video editing software, as it does well for audio recording too and is free to use. There were a few challenges here. Before importing the tracks as well as after exporting the final versions, I would run them through a program called CN Levelator, which is a free tool to level audio tracks. This has been a lifesaver for me in the past and absolutely was needed here too since Steph and I had very different mics. The other issue was syncing. Although Steph and I were synced through Zencastr, we were still hearing each other a bit lagged, so there were plenty of bits of dialogue where we had to speak together where Steph was about a second behind me (she was a pro ignoring me being a second behind her during recording). This was easy enough to fix by dragging my dialogue to be more in line with Steph’s.

I would listen to both takes and then pick the one that overall felt the best. Then I would clean up the audio, taking out awkward pauses and bits where we mucked up. After that I would mix and match bits of dialogue from the other take where I felt it was just a bit better and could punch things up a bit. Since the tapes were only 2-3 minutes each, it didn’t take me too long. As I went through the process, it went from about 40 minutes of editing per tape to 25 minutes per tape.

After I had the vocals roughly edited, I needed to figure out where sound effects might be needed. Some sound effects I had recorded live with the actual props. Others, like a door closing and a beeping device to sound like morse code, I had to search for through royalty free sound libraries. I think I spent a good two hours listening to beeps before I found one that was somewhat suitable. Yes, two hours is too long. I layered the sounds in with the dialogue, adjusting the timing of lines as needed. It was pretty amazing how much I was able to tighten up the dialogue just by playing around with adding or removing pauses.

I created separate files for the playlists, inserting ghostly whispers from Steph near the beginning of key songs as part of the final puzzle. I had been thinking of layering her voice overtop, but then didn’t want to take the risk that the player could not hear what she was saying over the music.

I had my vocal tracks. I had my music playlists with ghostly whispers. I was ready to get these on tapes.

And then…I lost my sanity.

So it turns out making a mixtape is different from recording scripted dialogue onto tapes. If the music playlist sounds a little crappy? Not great, but eh, you end up kind of ignoring it. When it’s just two people talking though? Good grief. I downloaded the vocal dialogue onto my phone and played them next to the tape recorder.

It sounded terrible.

I tried the same with the Google Nest, playing the vocal tracks right next to the tape recorder. I tried it with one tape recorder.

It sounded terrible.

I tried switching the tape recorders since I had two with recording capabilities.

The second tape recorder constantly stalled.

I tried putting the tape recorder and the speaker in a blanket fort with all my heaviest blankets to attempt to block out all that background static.

It made no difference.

This was bad. This was very bad. My main mechanic, the whole reason I wanted to make the game, was just…not…working.

Thus began a deep dive into the world of cassette tapes and cassette tape recording that I probably should have done at, oh, I don’t know, the BEGINNING OF THE PROJECT. Despite growing up with tapes, it seems I know very little about them. I turns out what was working against me was the quality of tapes themselves. You see, I was working with Type 1 tapes, the kind that is available to the average consumer. If I actually wanted the tapes to sound good, I would need something like Type 4 tapes, which is made of metal rather than oxide but which is far more expensive.

Well, I knew that I would not be shelling out any more money for this thing. I began considering the possibility I would need to simply send digital files directly to my Secret Santa. The problem with that is that I had signed an agreement with Steph that I would not share the files digitally since there are risks that her voice could be stolen and be replicated using AI (a very real danger to actors these days, I was glad to sign). If that was the case, I would need to talk to Steph about possibilities.

Then, after looking up a few other things I won’t go into, I wondered if it was somehow possible to record to a tape player directly from a computer. And…there is. Don’t get me wrong, I had to sift through a lot of sites where the solution involved “Hook it up to this incredibly complex and expensive tape recording system”, but eventually I found what I needed: if you can get a cable that has a 3.5mm jack on either end, you simply plug one end into the headphone jack of your computer and the other end into the microphone jack of your tape recorder. Start the recording on both machines and….that’s it.

That…was it…at first I began to go to Amazon to find the cable I needed but realized, I already had one! My Bluetooth headphones had come with a cable in case I wanted to hook them up manually to a device. And both ends are a 3.5 mm jack. Hands shaking slightly, I hooked up the tape machine to my computer as instructed, pressed record on the tape player, played one of the tape tracks, rewound the tape after and pressed play…

It sounded terrible.

Initially, there was a lot of swearing and despairing until I got a hold of myself and realized this was a different sort of terrible than the other terrible playbacks I had heard. This sounded…distorted…like when you play music too loud. I wonder…could I control the volume of the recording by lowering my computer volume? I repeated the process with a much lower volume, rewound the rest of the tape and…

It worked…

IT WORKED! Of course there was still some static, but that added to the nostalgia and it was a vast improvement from my previous results. It was clear. It felt haunting. I kid you not, I wept a little in relief. It would work.

I proceeded to transfer the recordings to the tapes in two steps: first, I transferred the music playlists. After all, in the narrative, the tapes were originally music mixtapes. Then, I recorded the vocal dialogue overtop of the tape, making sure I gave myself enough space that it didn’t sound like it was cut off at the beginning or end (unless designed that way). I was super happy.

I learned a LOT in this process. Most importantly, that if ever I do this again I do my research BEFOREHAND, not during the process. I spent a lot of hours agonizing over how to accomplish this. At the very least, if I work with cassette tapes again, and I will, I at least know what to do.