The job of the mic is to convert the sound waves generated by your voice from differences in air pressure to differences in electrical voltage. It is the electrical equivalent or analog of the sound waves and there is a strong analogy with the way the human ear works.
But we are dealing with digital sound, that is to say the sound recorded is converted into a digital format which makes subsequent post processing so much more convenient. It’s important to have at least an appreciation of the basic principles as there are some decisions you might make which will have ramifications for your podcast.
The digitisation of the audio that your microphones pick up is achieved by a process called “sampling”. The analog electrical sound wave generated by your mic is “photographed” or sampled many times each second and converted into digital information. These two factors will determine the quality of the digital audio file that is produced: the greater the number of samples per second and the greater amount of digital information ascribed to each sample, the higher the quality of the resulting file.
The number of samples per second is measured as frequency or Hz. If we look at CD quality sampling then the analog electrical signal is sample 44,100 times each second (44,100Hz or 44.1kHz)! Each of these samples or snapshots are described digitally as one byte of data, each byte comprising of 16 bits.
So each second the electrical sound wave is represented digitally as 44,100 x 16 bits of data, that’s 705,600 bits or 705Kbits or 0.7MB. So a 5 minute (300 second) recording in CD quality is 44,100 x 16 x 300 = 211680000 bits or 211MB, which is a pretty sizable file.
By way of comparison, the basic “uncompressed” digital voice telephone call is sample 8000 times a second with each sample being assigned 8 bits per byte which gives a “bandwidth” of 64,000Kbps.
The ZoomH4N can record in different qualities, in one particular mode it can record in a quality that exceeds that of CD with 96,000 samples per second (96kHz) each sample generating 24bits per byte with resultant massive files. A 300 sec recording in this format will generate a file of nearly 700MB!
If your podcasts are going to be of any substance they will need to be several minutes long - a sound bite will not be able to convey enough information to go into any subject in any detail. Typically audio podcasts are about 30 minutes in duration. You can imagine from the process described above, just how big the recordings will be!
Fortunately that’s really as much as you need to know about this side of podcasting. You can record in very high quality, higher than CD quality, but the result is an increase in size of files.
This means that you need to have sufficient disk storage to store them and have a sufficiently powerful computer to manipulate them when editing them. Also you will have to make sure that the memory card which is used by the H4N to record onto has sufficient capacity to capture the recording that you are going to make.
However something else that you need to be aware of which is related to this is something called compression.
Clearly, to transfer a file of nearly a Gigabyte in size would require a lot of bandwidth and storage capacity. the 16GB iPhone I have would be able to store only 16 x 1GB podcasts! Although the quality of the recording may be wonderful, the quid pro quo of long download times given the poor state of current bandwidth on the internet, and large storage requirements, outweighs the benefits of this super-quality.
So there are ways to make these files smaller for distributing to your audience which makes them a lot more convenient to deal with. Possibly the most widely recognised compression format that has impinged itself on the general public’s consciousness (without them necessarily realising what it might actually be) is that of MP3.
The world of compression and file formats is a complex area and I don’t intend to go into any great detail as it is beyond the purpose of these blogs. But it is important to have at least an appreciation of the main points.
MP3 makes files smaller by removing a lot of the digital information that defines the content, relying on the inadequacies (or the brilliance) of the human ear to “get away with it”. Many podcasts are distributed as MP3 and many of us may have our music stored as MP3 files. The trade off here is a decrease in quality of the audio and limited options for what else you can do with the podcast.
In my opinion a better format to compress to and distribute is AAC. AAC removes less information from the uncompressed recording, so the file sizes are larger than MP3, but on the other hand the quality is superior and much harder to distinguish from a CD quality recording.
It’s all about compromise.
So a half hour two person interview audio podcast recorded in CD quality might be approximately 300MB in size, this will be compressed as an AAC file to around 60MB, something that is much more manageable.
Just before we leave this section, the two pointy objects on top of the H4N are microphones. It comes complete with a pair of mic which are cardioid in nature and which are good quality. So why do I suggest using tie mics?
You could record perfectly good and usable audio podcasts using these in-built mics but again, you need to be aware of the implications. Because these “in-built” mics won’t be as close to the sound source as the tie-mic, they will capture much more of the atmosphere of the room or location in which you are recording. This means that if there is the background hubbub of a break out session at a seminar, or if the room is particularly large with hard echoey walls and floors, then these attributes will be captured much more so than with the in built mics than with tie-mics which by their nature are closer to the source of the audio.
Also, by using the H4N as a handheld recorder you run the risk of the mics picking up the noise of your hand on the casing of the device.
The next subject to consider is the structure and content of your podcast which we’ll look at in the next blog.