Not every project that I tackle comes to fruition, and I’m forced to abandon more than a couple (like the tracking motor to use with the eclipse, for instance.) But, as the title says, this one did.
For the record (Hah! I keel me!) this was the previous incarnation of it mentioned within the video. But here’s what I was dealing with this time around:
I have a decent audio recorder, and a decent lapel microphone. But the mic is intended for use with smutphones, and so has the quad-pole connector that those use, which allows for earphones and phone controls. This does not play well with the audio recorder designed before those were in constant use – that instead wants a tri-pole connector (stereo.)
I still have the little microphone seen in that linked post, but its audio quality is not very good and produced too much residual noise, more than I really wanted to edit out (especially since, with ongoing ambient sounds, there was no way to record a patch of ‘silence’ as a baseline to subtract this residual noise.)
So I switched to using the nice lapel mic with the smutphone as a recorder, but then went through three different audio recording apps before settling on the fourth, since the others wanted to do their own thing with auto-levels, essentially deciding what the base volume (okay, gain) should be. This produced some really terrible effects, given that the ambient sounds could get quite loud, especially in the height of cicada season, and quite variable, with the goddamn traffic too close by. What this produces is a lot of warbling and vague underwater-like sounds, very unrealistic and distracting.
Those ambient sounds were quite a chore in themselves, since at times the cicadas really started winding up, and diesel trucks were passing, and so on. The day I chose for the video above was the quietest of all my attempts, and you can hear how un-quiet it really was.
And then, once I got decent audio, it was time to synchronize it with the video – which came with its own audio track that was quite useful in itself, though it captured absolutely none of the nest sounds. Meanwhile, the nest mic captured too little of the ambient sounds, making the adults seem incredibly distant. That’s what they’re made for of course, to record the wearer and not everyone else, but it did mean that I wanted to keep both audio tracks.
Since I simply set the smutphone to record in its position underneath the nest and left it there, it produced one long recording track, while I was able to start and stop the video camera as appropriate to the action and lack thereof. This meant that I had four video tracks of two to five minutes and one audio that was eighteen minutes long, with ambient noise none too distinguishable throughout. You know those little clappers that you see right before they shoot a movie scene? That’s what those are for: syncing separate video and audio tracks. Without such a thing, I was forced to sync them by finding unique sound fragments that appeared in both recordings, such as some other birds and distinctive passing vehicles. Note, however, that these largely had to be down to fractions of a second, or curious ‘echoes’ would be produced by the sounds that carried onto both tracks but weren’t perfectly aligned. There may be methods to jog the track alignments by milliseconds in the editing program that I used, but if so, I didn’t know where it was, so I was doing it by twitching the mouse – a bit tedious.
Once these were aligned, then I had to trim them down, ensuring that all tracks were cut in the same place and stayed aligned. Add in selecting the right gain for each (boosting the nest audio slightly, reducing the videocamera’s by an almost-equal amount.) Once satisfied with the running edit, then I had to record the voiceover track on another program while watching the video, clean that up, insert it into the video editing program, and then play with the gain and synchronization on that. Oh, yeah, and add in the title and close images.
And then, upload it to Vimeo and do all the little doodads that that requires. This means a five-minute video can take several hours to get together – be nice if it better reflected all this effort, right? Keep your commentary to yourself.
So while I like including video and know that this comes with its own learning curve, sometimes this drags on a bit, you know?
* * *
For the record, I am presently using Kdenlive for the video editing, and Audacity for the audio – both extremely capable programs for completely free software. For the nest recording, this was through ASR Voice Recorder from the Google Play Store (yes, Android.) But I also can’t stress enough how much easier it is when your microphones are decent. The lapel mic was a solid find from some generic source, far less expensive that its capability reflects, while the main recording mic on the desk (voiceovers and podcasts) is a Samson G-Track – expensive, but I picked it up used and it’s been superb.