tom's web

Making Sound Messages
by tom

 

Disclaimer:

Please remember that my experience with sound-software is limited. There are probably many more options available;  more places to go to get software;   and more things to do than I realize. What I have written down, below, should get you started and perhaps, warn you of some pitfalls.

Note: "PureVoice" and "PVConv" are (free) programs obtained from:

http://www.eudora.com/purevoice/pluginDL.html

Notation

sound-message or sound file

.. is the sound equivalent of a test-message and text-file.   Sound-files can be sent over the Internet to a friend's eMail address (as an attachment) .. just as normal eMail is sent.

At the receiving end, the user must have sound-software, sound-card, and sound-speakers to hear sound-messages.  Sound-messages are easy to make. All that is required is a microphone connected to the sound card and MS software to record a sound-message .. but .. read on.

In many ways this sound-message making process is similar to making a text file .. that is .. a keyboard(microphone) is used to type-in a text(sound) message which is then seen(heard) using a screen(speakers). Both can be sent to an eMail address.

.qcp

.. is a sound-file name extension used by PureVoice for their efficient compressed and filtered sound-file format. However ".qcp" belongs to PureVoice and is a non-standard file extension/type.

.wav

.. is a standard sound-file name extension used by MS Windows and others. "Wave" or ".wav" files are easily played with Windows software.   However, ".wav" files are very large and consume large amounts of storage and "bandwidth."

filtered

.. means to eliminate sound frequencies which are (considered) unnecessary to the understanding of a sound-message.  "Filtering" is done to reduce "bandwidth" (as much as possible) and is a common practice in telephony.

compressed

.. means the simple elimination of redundant bits/bytes. For example a series of repeated spaces can be coded as just one space during compression .. then expanded, later. Again the purpose is to reduce "bandwidth."

... back to top

Discussion

Note .. my file size estimates (below) should be treated as "ball-park" approximations, only

These notes are about *two* methods that can be used to make sound messages. The first level describes a quick and efficient way to make a sound message using "PureVoice" and a microphone. However, this (free) software does not allow you to edit or enhance your sound-messages.

The second level is more complicated and requires additional software support such as a sound-editor (e.g. "CoolEdit") and file converter (e.g. "PVConv).

Sound messages can be created and sent, simply, using "PureVoice," and a microphone, only

.. that is .. much of what I have written, below, is unnecessary if all you want to do is to simply and quickly make a sound-message to send to a friend.

Level 1 (making .qcp files only):
"PureVoice" is used to directly record a sound-message using a microphone. The resulting sound-file can, later, be sent over the Internet or played back using speakers. The microphone and speakers are connected to the computer's sound card. "PureVoice" creates an efficient sound-file low in bandwidth. Typically, "PureVoice sound-messages are about 1 minute (30-80K bytes) in length so Internet transmission time short.

***note*** if you use level 1 .. then skip all of level 2 notes

Level 2 (making enhanced .wav files to .qcp files):
Windows 95/98/NT software comes with software that allows you to directly make a ".wav" file using a microphone connected to a standard sound card. However, ".wav" or "wave-files" are quite large (approx. 1-meg per min.) and consume bandwidth and, therefore, do not lend themselves to extended messages such as complete musical songs.

To make a "wave file," connect a microphone into your sound card;  click on .. <Start> <Programs> <Accessories> <Multimedia> <Sound Recorder>;   say your message;  and then save the result as a ".wav" file

Alternately (.. and preferred)
a sound-editor program (such as) "CoolEdit" allows you to record a voice message directly, and mix it with music section that has been recorded from a CD (or other sources). In addition, a sound editor allows you to *easily* edit (delete, insert or enhance) portions of the sound file.

"sound editing" is done using a visual display of the wave pattern and is quite easy, direct and accurate. Some sound editing features are: deleting "dead" sound times; inserting a voice track;  increasing/decreasing volume in selected portions of the wave form; decreasing noise level;  and many other built in sound enhancement and effects.

Because ".wav" files are large and consume bandwidth, PureVoice also makes an effective sound conversion utility program which filters, compresses ".wav" files into ".qcp" files. For example,  a 1 meg. ".wav" files can be compressed down to about 80K .. which is about the size of an image file.

Because the "PVconv" conversion program is not built into the (free) software program, conversion must be done manually, as a separate step, with the "PVConv" program.

e.g.      PVConv     xxxx.wav    yyyy.qcp

where the user supplies the input (xxxx.wav) and output (yyyy.qcp) sound-file names.

***note*** arguments are required for the "PVConv" command

Once "PVConv" is started it takes time to run .. and gives no user-friendly messages while it does the conversion. The only run-response that you see on the screen .. is that the cursor appears to "hang" on the next screen-line after the <Enter> is pressed.  Just Wait .. until the processing ends.

to stop the processing enter a <CTRL-C>

... back to top

Requirements

Hardware:

Software:

Optional:

... back to top

Comments

The PureVoice software components are not user friendly .. each time you run PureVoice or PVConv they return to their default setup options .. rather than retain their last values.

CoolEdit (.. and there are others) seems to have all the basic sound editing features needed to edit a voice message, such as:

  • eliminate noise
  • change amplitude
  • cut out sound sections
  • record-over voice sections
  • insert other sound file/s
  • many more sound enhancement features

"CoolEdit" is compatible with many types of sound-files .. and allows you to convert from one to another.

"PureVoice" recognizes "WINPCM" type of ".wav" file.  only

These notes will assume that any given input sound-file can be converted into a "WINPCM" type (step #2, below) using your sound-editor

These notes do not discuss how to run the sound editor or edit a sound-file.

... back to top

Procedure
(Making Level 2 Sound Files, only)

0) General setup:

First create a "$PureVoice" directory to hold *all* sound-files .. organize by types

i.e.  such as voice-messages, music, sound-clips .. etc.

 

Overview .. the procedure describe below does the following:

  • take any input sound-file .. and convert it to a WINPCM, ".wav" file (step #1)

  • enhance/edit the WINPCM ".wav" file (step #2)

  • save the output sound file as a WINPCM ".wav" file (step #3)

  • convert the WINPCM ".wav" file to a ".qcp" file (step #4)

  • play the ".qcp" file using PureVoice software (step #5)

Now Let's do it ...

1) Next .. run "CoolEdit"

to (edit and) convert a sound-file to a WINPCM ".wav" sound-file.

to setup initial sound file parameters

<File>
<New>    ( to establish the input/output sound-file parameters)

<OK>

2) to open and convert input sound-file to a WINPCM ".wav" file

<File>
<Open As>

<OK>

Next .. check the wave form displayed on the screen .. does it look right? .. any error messages?

Next .. edit .. delete, insert, enhance etc. .. as needed

3) to save the (enhanced?) sound-file as a Windows PCM ".wav" file

<File>
<Save As>

4) to convert the .wav-file to a .qcp-file  using the "PVConv" program

run the "PVConv" program ( .. in DOS or whatever)

  e.g.   PVConv   xxxx.wav   yyyy.qcp    
         (include *both* the input and output file name w/extension

5) to check the results

run "PureVoice" program and listen to the results ..

  • If the sound clip sounds too slow .. do again and decrease "sample rate" downward.  Normally 8000 is about right (see step#1)
  • If the sound clip sounds too fast .. do again and increase "sample rate" upward. Normally 8000 is about right (see step#1)

6) last .. to verify the results

.. by sending this .qcp-sound-file to yourself .. as an eMail attachment.

 

Further Suggestions:

   http://www.purevoice.comm

... back to top

-- End --