Audio for Streaming: A Content Creation Scenario

---

December 10, 1998

• Introduction
• Background
• Recording a New Sound from Videotape
• Audio Editing
• Audio File Preparation

Introduction

Just as with images and video, depending on your hardware requirements and budget, you can produce a range of sounds from beeps and clicks to full musical scores for your presentations using Windows Media Technologies (WMT). You can find audio software as freeware or shareware, or you can pay thousands of dollars to get professional audio recording and editing capabilities. This article will not cover high-end professional audio production, but will instead offer a scenario to help point you in the direction of creating quality audio economically and easily.

The example covers:

• Recording a new sound from videotape
• Audio editing
• Audio file preparation

Note   The following example was originally published in the Microsoft® Windows Media Content Creation Authoring Guide and is included in the Windows Media Technologies JumpStart CD.

Several months ago, I made a video for Human Resources recruitment efforts. For this video, a professional music sound track was created. The project I'm working on right now is to develop an illustrated audio presentation using Windows Media Tools. Simply put, illustrated audio is a synchronized audio track with images, text captions like closed captioning on TV, and, if desired, different VBScript or JavaScript features, depending on how much scripting you want to do.

Recording a new sound from video tape

1. I already have the images I'll use; I just need some cool background music to listen to during various parts of the presentation. The sound track on our recruitment video would be exactly what I'm looking for. I just have to figure out what I actually need for the illustrated audio project and then how I get the audio from videotape to the format I need.
2. First, I have to determine what type of audio file I need for the final presentation, the browser, and the bandwidth of the network over which the content will be played. The browser and bandwidth issues are easy; this is for internal use in our company and IE 3 is our corporate standard. The network issue is determined for me; our information services group allows delivery of content up to 100 Kbps. The WMT server acts as a gateway for this point; the server can be set up by an administrator to deliver content only if it is less than a predetermined data rate.

Audio Editing

1. Because the audio track is on videotape, I have to digitize it from the tape source. This can be done in two general ways, one being much better than the other. The first way is to merely play the videotape and record the sound using a microphone plugged into my sound card. Sure, this is easy, but the quality would be terrible. I'd get too much background noise from the air conditioner, people talking in the hall, and so on. So that's not a viable option.
2. The other method, and the one I'll use, is to record the audio directly from the videotape into the sound card. This way I get a sample as close to the original as possible using a standard computer sound card and a VHS tape deck.
3. I won't go into the exact cable hookups between the VCR and the sound card, but basically the left and right audio-out terminals on the VCR are connected to the line-in or microphone terminal on the sound card. The next step of setting up and balancing the recording volume actually is the best determining factor for which sound card connection to use.

Audio File Preparation

1. Balancing your recording levels means optimizing the recording volumes so that you get the highest possible distortion-free volume without getting extraneous background noise from other devices.
2. Open the Properties dialog box for the sound card in your machine and select the recording properties. This should display the master volume recording panel. Some trial and error is involved, so for starters, place the master volume level at mid-range and set the line-in or microphone level near the maximum setting.
3. You might see other devices, such as CD-ROM or Synthesizer (MIDI), in the recording panel. Mute all non-essential devices; this cuts down feedback and background noise.
4. Now open the sound recording program that you'll be using. One of the simplest is Sound Recorder, which comes with the accessories when Windows is installed. More sophisticated sound recording packages, such as Sound Forge from Sonic Foundry, should be used if you want to record very long pieces of audio. Sound Recorder records the audio to the memory (RAM) of your computer, so unless you have LOTS of RAM you'll only be able to record short audio clips. Look for an audio recording and editing package that allows you to record to disk so that you can capture much larger pieces of audio.
5. Start the playback of the VCR and select the Record button in your sound recording program. You should see a wave pattern display of the audio as it is playing. What you want to do to balance your recording volume is look for the highs and lows of the audio as it is being recorded. If the volume settings are too high, you see the wave patterns being clipped off and appearing to fill the recording window. If the settings are too low, you see only very small variations in the wave pattern.
6. If the sound is being clipped off because the volume is too high, move the line-in or microphone volume setting down until most of the volume range is displayed without clipping.
7. If the sound is too low, increase the volume setting. Ok, here's the trick with the line-in or microphone input. Depending on your sound card and the volume of the audio source, you might not get enough volume using the line-in input. In this case try the microphone input, as the volume sensitivity through the microphone input is generally greater than with the line-in input.
8. I only had to make minor volume adjustments with this audio source. Generally, if the audio has been professionally recorded, part of the recording process is to balance the source and get maximum volume with minimal distortion. This simplifies the recording process on my side, as I don't have to try and adjust for wide swings in volume. Keep this point in mind when you're recording audio that will be source material for some future work.
9. In my sound recording program, set the recording options for the highest-quality sound possible; with most sound boards this is 44.1-kHz, 16-bit stereo sound. I'll need to change this for the audio used in the illustrated audio, because this high-quality audio would be far too large to deliver over our corporate network, not to mention the Internet bandwidth.
10. OK, let's rewind the tape and start the playback from the point I want to record. Hit the record button and stand by. When finished recording, save the file immediately. Remember the last time you'd didn't save the file?
11. In the Save dialog box, many different options are available depending on what codecs I've installed previously. At this point, I want to keep the audio as high-quality as possible. I'll select the uncompressed attribute, which saves the file as PCM, with the same options used for the recording. In this case, 44.1-kHz, 16-bit audio.
12. Now that I have my digital source file, I need to do some editing and then compress the file into a final delivery format. I'm going to use Sound Forge to edit and compress my file. So I reopen the file and first play it back to see that it sounds like I expect it to.
13. There are a few clicks, pops, and other artifact noises that I want to get rid of. Sound Forge and some other robust audio editing tools have the capabilities of analyzing the audio file and removing extraneous noise. After I've done this and put a fade in and fade out at the beginning and ending, I'm ready to compress the file and degrade it to a bandwidth closer to what I'll need in the final piece.
14. I save the file as 22-kHz, 16-bit stereo using the MS ASPCM codec. This does some compression, but retains very high-quality sound. The file is probably still too large for my illustrated audio, but I'll use the Microsoft WMT T.A.G. Author for final determination of a codec. If you'd like to get more information on the Microsoft WMT T.A.G. Author, take a look at the section on Creating Illustrated Audio or refer to the documentation on the Microsoft WMT T.A.G. Author that is available when you download the Windows Media Tools.
15. The audio file is basically ready for use in a specific project. After this it is time to move on to the next steps of making the illustrated audio. See Converting Digital Audio and Images to ASF.