Slicing and Dicing

As we covered last time, the Bible is a big book. Once we’ve recorded all of our audio, we’ll still need to get those 177+ gigabytes of data onto a smart phone, and potentially all delivered over a slow 3G network. How on earth will we be able to do that?

The solution is to slice up the audio into the pieces we need, stitch them together, and send just that little chunk of data to the phone.

Let’s take a hypothetical listening plan as an example. We’ll call it Nuggets of Wisdom. Each day will include a few verses on a related topic, pulled from across scripture. Here’s what Day 1 looks like:

Day 1: “Faith”. Hebrews 11:1, Romans 1:17, Mark 9:24, Mark 5:36, 2 Timothy 4:7

(A listening plan that was actually this scattershot would probably be terrible, but it’s a helpful example, so let’s go with it.)

Looking at the verses for this day of the plan, we’ll need some of the audio from five chapters: Hebrews 1, Romans 1, Mark 9, Mark 5, and 2 Timothy 4. We have the recordings for each of those, but how will we extract only the verses we need? We’re going to need to know where each verse starts and ends within the audio of the full chapter. For that, we’re going to need another tool.

Enter the Verse Offset Marker Editor

Using this tool we are able to create a marker for each verse boundary in a chapter, and save that data to our growing database.


This means we have to set these markers for every chapter in the Bible, read by every voice artist. It’s a lot of work, but it makes our entire platform possible.

Once we have that data saved for the chapters we need, we can then use our audio toolchain to extract just the portion that we need from each Master Recording, then stitch them all together into a single file. We use a tool called sox to do this.

So now we have just one big wav file with all of our audio slices spliced together. But it’s still too big. We need to make it smaller. This is where our old friend mp3 comes into play. We convert the audio file from wav to mp3, and store it on our servers.


Now, when a listener is ready to hear the listening plan for Day 1, all they have to download is a small audio snippet for that morning, easily downloadable on even a slow 3G network.

Just like the old proverb, How do you eat an elephant? One bite at a time.

Jeff McFadden