Digitised speech for the internet

1.   Recording tapes to computer

Connections

You will need a lead to connect your audio to the computer input.  Don't buy an expensive sound card or leads if you are only dealing with speech. The sound kit which came with your computer is good enough.

Maplin or similar shops have all the leads you might need.  The computer end is almost certainly a 3.5mm stereo mini jack.  The other end is most probably either twin (left and right) phono plugs (stereo system) or 3.5mm mini jack (Stereo for walkman type gear or monofor cheap old portable tape recorder).

You need to connect to "tape out", "line out" from the tape machine.  An alternative is to use the headphones socket (may be a large stereo jack in that case).

If your sound card has a "line in" input then connect to that. These can be hard to identify sometimes. Modern connectors have a blue plastic surround which makes it easy.  The symbol may be an arrow pointed out of the  centre of three concentric circles (v easy to confuse with line out which has the arrow pointing the other way) or may have the words "line in" stamped on the metal.

If there is no "line in" (unlikely) then you will have to use the microphone input instead (most laptops).

Recording Software

There is plenty of choice here.

http://www.nctsoft.net/index.htm

Has “Power Audio Editor” which is a later version of what I use.  This does more than you probably want but is a good resource for converting between file formats etc as well as recording.  It is also well up to the task of recording music to a high level too if your musicians want to do that.  I use it to record BBC online radio progs from time to time if I want to keep them.

There are lots of other choices or your can use the built in sound recorder which comes with windows. This is limited to recording 1 minute by default but there is a work around see

http://support.microsoft.com/kb/82215/en-us?FR=1&PA=1&SD=HSCH

(Microsoft have already moved this article once.  If they move it again go to the Microsoft knowledge base and search for “How to Increase the Maximum Recording Time in the Sound Recorder Utility”)

Make the recording

Trial and error rules here. The main thing is to have max gain (volume) without distortion.  This gives you the best signal to noise ratio.  i.e. more of the sound you do want – pearls of wisdom etc. – and less of the sound you don’t want – tape hiss, rumble, hum and whatever.

If the original recording used noise reduction (e.g. Dolby) turn the treble down a bit.

2.   Other recording methods

Straight to computer.

This works well, either through a mixing desk, from the line out of your PA system or a microphone. The built in mike of most laptops is not usually good enough.

Portable mp3 or iPod sound recorder

I am just experimenting with one of these but first results are promising.  You wear this little device round your neck.  Some plug straight into a USB socket on your computer.

3.   Processing the sermon for publication

The raw sound file will usually be too big for uploading to your web site.  If you recorded straight to mp3 format then it will probably be 20-60 mBytes in size and much bigger if it is a PCM/Wave file.  And, although mp3 players now have a large amount of memory, it is neater to trim the output to the size you need.  Windows sound recorder can convert files

Many modern CD/DVD players can handle mp3 format files now as well.  This means you can store around 15 sermons as mp3 files on a CD or many more if you take the trouble to compress them a little more.

For web publishing I recommend the “RealAudio” format (as used by the BBC).  Highly compressed mp3’s are not bad but they end up a bit larger (15-20%) and do not sound as mellow.

There is a free converter available from Real called RealProducer Basic.  Download from

http://www.realnetworks.com/products/producer/basic.html

when converting a sound file first choose your input file then click the button labelled audiences and choose these settings:

Audio Mode: Voice

Video Mode: None

Tick the box for high quality resampling.

Ditch any “audience” listed and instead choose “16k substream for 28k dialup”

Then close the box and choose the Menu Item “Settings” and save the current ones as default.  Every time the programme starts it will now choose the best settings for internet audio.

There are many alternatives for embedding sound in websites.  Copy my source code if you want.  Real player needs two files.  The recording itself (with the suffix .ra or .rm) and a little text file which points to the location of the recording (with the suffix .rpm).  There are tutorials for all this on the Real Networks web site.