From lojban-out@lojban.org Fri Jul 14 19:03:56 2006 Return-Path: X-Sender: lojban-out@lojban.org X-Apparently-To: lojban@yahoogroups.com Received: (qmail 46013 invoked from network); 15 Jul 2006 01:36:13 -0000 Received: from unknown (66.218.66.166) by m34.grp.scd.yahoo.com with QMQP; 15 Jul 2006 01:36:13 -0000 Received: from unknown (HELO chain.digitalkingdom.org) (64.81.49.134) by mta5.grp.scd.yahoo.com with SMTP; 15 Jul 2006 01:36:12 -0000 Received: from lojban-out by chain.digitalkingdom.org with local (Exim 4.62) (envelope-from ) id 1G1Z1g-0001Vd-Ei for lojban@yahoogroups.com; Fri, 14 Jul 2006 18:33:00 -0700 Received: from chain.digitalkingdom.org ([64.81.49.134]) by chain.digitalkingdom.org with esmtp (Exim 4.62) (envelope-from ) id 1G1Z0y-0001VB-FQ; Fri, 14 Jul 2006 18:32:18 -0700 Received: with ECARTIS (v1.0.0; list lojban-list); Fri, 14 Jul 2006 18:32:08 -0700 (PDT) Received: from nobody by chain.digitalkingdom.org with local (Exim 4.62) (envelope-from ) id 1G1Z0V-0001Ul-W6 for lojban-list-real@lojban.org; Fri, 14 Jul 2006 18:31:48 -0700 Received: from nz-out-0102.google.com ([64.233.162.192]) by chain.digitalkingdom.org with esmtp (Exim 4.62) (envelope-from ) id 1G1Z0T-0001Uc-La for lojban-list@lojban.org; Fri, 14 Jul 2006 18:31:47 -0700 Received: by nz-out-0102.google.com with SMTP id l8so266803nzf for ; Fri, 14 Jul 2006 18:31:44 -0700 (PDT) Received: by 10.36.33.3 with SMTP id g3mr68061nzg; Fri, 14 Jul 2006 18:31:44 -0700 (PDT) Received: from blackbeast ( [24.8.249.59]) by mx.gmail.com with ESMTP id 6sm66568nzn.2006.07.14.18.31.43; Fri, 14 Jul 2006 18:31:44 -0700 (PDT) Date: Fri, 14 Jul 2006 19:34:57 -0600 MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_001F_01C6A77C.97CB1550" X-Mailer: Microsoft Office Outlook, Build 11.0.5510 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2869 Thread-Index: AcanruGKG9CRLrQQT7i0t51G8WDECg== Message-ID: <44b84580.401594d5.016c.0f63@mx.gmail.com> X-Spam-Score: -1.5 (-) X-archive-position: 12210 X-ecartis-version: Ecartis v1.0.0 Errors-to: lojban-list-bounce@lojban.org X-original-sender: matthew.dunlap@gmail.com X-list: lojban-list X-Spam-Score: -2.1 (--) To: lojban@yahoogroups.com X-Originating-IP: 64.81.49.134 X-eGroups-Msg-Info: 1:0:0:0 X-eGroups-From: "M@" From: "M@" Reply-To: matthew.dunlap@gmail.com Subject: [lojban] lojban talks, xu le skami cu tirna X-Yahoo-Group-Post: member; u=116389790; y=9Kr-hEMRLk8zw8IjLkKr7nMS1ZZGD_rAyTRpHmvXqKzic3oICw X-Yahoo-Profile: lojban_out X-Yahoo-Message-Num: 26637 ------=_NextPart_000_001F_01C6A77C.97CB1550 Content-Type: multipart/alternative; boundary="----=_NextPart_001_0020_01C6A77C.97CB1550" ------=_NextPart_001_0020_01C6A77C.97CB1550 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit So a while ago I promised everyone a quick and easy guide to making computer audio recordings sound good enough for indie movie use. Now that we're entering the translation phase of L&P I suppose now would be as good a time as ever to release it. There are only two important aspects to good recording; picking up the signal, and not picking up any other noise. Everything else revolves around that Signal to Noise Ratio (SNR). Once the audio is successfully digitized many things can be done to modify it, however, for the sake of a consistent product from a distributed talent base it is best not to attempt any digital audio editing and let one person who is acting as audio editor for the whole project do all of the work. Thus, this guide is only about getting a high quality (that is, a good SNR) signal into the computer. The most obvious sources of noise (where noise is anything audible that isn't the desired signal) are obvious: Anything that moves Centralized-air comfort systems People elsewhere Communications (telephone, IM, cell phone, neighbor's dog, etc) Computers (especially the one doing the recording) For the most part, simply disabling/stopping/removing/otherwise-silencing the noise maker is enough, but the case of the computer being used as the recording medium is a little trickier. Lucky for us there is radio shack, where $7 is enough to buy a 1/8" miniphono extension cable. To eliminate the noise from a computer (quiet as it may be, the mic will still pick it up, and it will detract from the SNR) simply extend the microphone cord such that the recording can be made in a partitioned off area (optimally another room). Additionally, laptops often run with their fans turned off or on low speed with occasional bursts of high speed for cooling, if you have a laptop and decent timing (wait for high speed to end then hit record) the issue can be solved without any extensions at all. Other harmful things to the SNR are less obvious, but for the most part they're just as easy to fix: Gain too low Gain to high TPS problems Echoes RFI Gain is the % of possible volume that the computer regards various input signals as. If the gain is too low the recording will sound quiet (ie low signal) and require normalization, which will amplify the background noise as well. If the gain is too high the microphone will supply a signal larger than the computer regards as 100% volume and clipping will occur (think plateau instead of rolling hills). A good example of this throughout the beginning of jbocradi episode 1. The key to correctly setting the gain is rehearsal (which is a good thing to be doing anyway), open Audacity (yay open source) and click the recording level meter, this will give you real time feedback about the sound levels coming in from the mic. Now, rehearse. The loudest part of your dialogue should bring the meter to between the -6 and 0 marks. If it exceeds 0 the gain is too high, and audacity will alert you to this fact by keeping the far right bars (just past the 0 mark) illuminated red. To reset this meter simply click it twice and try it again with a lower gain (the adjustment bar beside the mic icon). If the meter never gets above -6 turn up the gain. TPS reports are the sharp explosive noises made by the letters T, P and S. Think about a time you've overheard whispering from the next room, you only hear the TPS reports, everything else is too quiet. Ordinarily your brain just cancels the extra volume, but microphones aren't that smart. The beginning of jbocradi ep 1 has a few TPS problems. In reality this is the exact same problem as having too much gain, but since these letters are so much louder, if you dial down the gain to accommodate you end up losing over all quality because it will later have to be equalized back up. The solution is a wind canceling microphone attachment. They can be had for $3 at radio shack. They're large and black, and look ridiculous on small mics (see the pic of mine), but they work miracles for TPS canceling. Installation is simple, especially if you're up to speed with modern prophylactic techniques. Also, B causes TPS reports, but that wouldn't have been as funny. Echoes are familiar to everyone, and they're easy to get rid of, just make sure the room that you have extended the microphone into is large and full of oddly shaped things (preferably soft ones). Sofas, beds, tables, pillows, desks, and shelving units make excellent echo cancellers because they break up the planarity of echo-causing walls. RFI stands for Radio Frequency Interference, and can cause all kinds of weird noises to be recorded that the person never hears. I'll spare you the physics of why this happens, but the tip to be garnered here is don't have a cell phone in the same room as you're recording, even if it's set to silent and vibrate has been turned off. If it is activated by someone calling it, it will respond by blasting a very "loud" RF signal back to the nearest tower, but in so doing it will cause a small current in the wires of your sound system which (once amplified) is enough to cause odd clicking/buzzing noises. Once you've taken care of all of these easy things, the next step is to go buy a nicer microphone, especially if you have a very deep voice like I do. I hope this guide helps! --teryret. ------=_NextPart_001_0020_01C6A77C.97CB1550 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

So a while ago I promised everyone a quick and easy guid= e to making computer audio recordings sound good enough for indie movie use.&nbs= p; Now that we’re entering the translation phase of L&P I suppose no= w would be as good a time as ever to release it.

 

 

 

There are only two important aspects to good recording; picking up the signal, and not picking up any other noise.  Everything else revolves around that Signal to Noise Ratio (SNR).  Once the audio= is successfully digitized many things can be done to modify it, however, for the sake of a consistent product from a distributed talent base it is best not to attempt= any digital audio editing and let one person who is acting as audio editor for = the whole project do all of the work.  Thus, this guide is only about gett= ing a high quality (that is, a good SNR) signal into the computer.

 

The most obvious sources of noise (where noise is anythi= ng audible that isn’t the desired signal) are obvious:=

Anything that moves

Centralized-air comfort systems=

People elsewhere

Communications (telephone, IM, cell phone, neighbor̵= 7;s dog, etc)

Computers (especially the one doing the recording)<= /o:p>

 

For the most part, simply disabling/stopping/removing/ot= herwise-silencing the noise maker is enough, but the case of the computer being used as the recording medium is a little trickier.  Lucky for us there is radio sh= ack, where $7 is enough to buy a 1/8” miniphono extension cable.  To eliminate the noise from a computer (quiet as it may be, the mic will still pick it up, and it will detract from the SNR) simply extend the microphone = cord such that the recording can be made in a partitioned off area (optimally an= other room).  Additionally, laptops often run with their fans turned off or = on low speed with occasional bursts of high speed for cooling, if you have a laptop and decent timing (wait for high speed to end then hit record) the i= ssue can be solved without any extensions at all.

 

Other harmful things to the SNR are less obvious, but fo= r the most part they’re just as easy to fix:

Gain too low

Gain to high

TPS problems

Echoes

RFI

 

Gain is the % of possible volume that the computer regar= ds various input signals as.  If the gain is too low the recording will s= ound quiet (ie low signal) and require normalization, which will amplify the background noise as well.  If the gain is too high the microphone will supply a signal larger than the computer regards as 100% volume and clippin= g will occur (think plateau instead of rolling hills).  A good example o= f this throughout the beginning of jbocradi episode 1.  The key to corre= ctly setting the gain is rehearsal (which is a good thing to be doing anyway), o= pen Audacity (yay open source) and click the recording level meter, this will g= ive you real time feedback about the sound levels coming in from the mic.  Now, rehearse.  The loudest part of your dialogue should bring the met= er to between the -6 and 0 marks.  If it exceeds 0 the gain is too high, = and audacity will alert you to this fact by keeping the far right bars (just pa= st the 0 mark) illuminated red.  To reset this meter simply click it twic= e and try it again with a lower gain (the adjustment bar beside the mic icon)= .  If the meter never gets above -6 turn up the gain.=

 

TPS reports are the sharp explosive noises made by the letters T, P and S.  Think about a time you’ve overheard whisper= ing from the next room, you only hear the TPS reports, everything else is too q= uiet.  Ordinarily your brain just cancels the extra volume, but microphones aren&#= 8217;t that smart.  The beginning of jbocradi ep 1 has a few TPS problems.&nb= sp; In reality this is the exact same problem as having too much gain, but since t= hese letters are so much louder, if you dial down the gain to accommodate you en= d up losing over all quality because it will later have to be equalized back up.  The solution is a wind canceling microphone attachment.  The= y can be had for $3 at radio shack.  They’re large and black, and = look ridiculous on small mics (see the pic of mine), but they work miracles for = TPS canceling.  Installation is simple, especially if you’re up to s= peed with modern prophylactic techniques.  Also, B causes TPS reports, but = that wouldn’t have been as funny.

 

Echoes are familiar to everyone, and they’re easy = to get rid of, just make sure the room that you have extended the microphone i= nto is large and full of oddly shaped things (preferably soft ones).  Sofa= s, beds, tables, pillows, desks, and shelving units make excellent echo cancel= lers because they break up the planarity of echo-causing walls.

 

RFI stands for Radio Frequency Interference, and can cau= se all kinds of weird noises to be recorded that the person never hears.  I&#= 8217;ll spare you the physics of why this happens, but the tip to be garnered here = is don’t have a cell phone in the same room as you’re recording, e= ven if it’s set to silent and vibrate has been turned off.  If it is activated by someone calling it, it will respond by blasting a very “= loud” RF signal back to the nearest tower, but in so doing it will cause a small current in the wires of your sound system which (once amplified) is enough = to cause odd clicking/buzzing noises.

 

Once you’ve taken care of all of these easy things= , the next step is to go buy a nicer microphone, especially if you have a ver= y deep voice like I do.

 

I hope this guide helps!

 

--teryret.

------=_NextPart_001_0020_01C6A77C.97CB1550--------=_NextPart_000_001F_01C6A77C.97CB1550 Content-Type: image/jpeg; name="mic.jpg" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="mic.jpg" [Attachment content not displayed.] ------=_NextPart_000_001F_01C6A77C.97CB1550--