RecAll (10)

MP3 Patent Countdown

Various Sagebrush programs are able to record and play MP3, but only if you already have an ACM MP3 audio codec installed on your computer, which was common during Windows XP but less  likely in later versions of Windows. We do not bundle an MP3 encoder with our products because the technology is covered by software patents, whose validity is recognized in the country where Sagebrush is located. The licensing fees are prohibitive.

When do the patents covering MP3 encoding and decoding actually expire in the USA? That is complicated, as issues relating to software patents tend to be. Tunequest provides a handy list of patents, with the last expiring December 30, 2017. The Wikipedia article on MP3 raises the question of validity of the later patents:

The various MP3-related patents expire on dates ranging from 2007 to 2017 in the U.S.[52] The initial near-complete MPEG-1 standard (parts 1, 2 and 3) was publicly available on December 6, 1991 as ISO CD 11172.[53][54] In the United States, patents cannot claim inventions that were already publicly disclosed more than a year prior to the filing date, but for patents filed prior to June 8, 1995, submarine patents made it possible to extend the effective lifetime of a patent through application extensions. Patents filed for anything disclosed in ISO CD 11172 a year or more after its publication are questionable; if only the known MP3 patents filed by December 1992 are considered, then MP3 decoding may be patent free in the US by September 2015 when U.S. Patent 5,812,672 expires which had a PCT filing in Oct 1992.[55][56][57]

Just for the sake of caution, let us take the more conservative date of December 30, 2017.

[ujicountdown id=”mp3patent” expire=”2017/30 23:59″]

(We are not lawyers, and nothing in this article should be taken as legal advice.)

We hope and expect to be supporting RecAll and RecAll-PRO well into the next decade, so it is quite possible to wait out this particular portfolio of patents.

Related Posts:

RecAll-PRO and RecAll on MacIntosh OS X

Previously we described how to run RecAll-PRO, RecAll, and other Sagebrush programs on Ubuntu Linux, using the free open-source Wine software library. Here we show how to do the same thing with MacIntosh running OS X on an Intel CPU.

The easiest way to install Wine on OS X is with WineBottler. Download and install the dmg file. Open the file and you should see this folder:

Drag and drop Wine and WineBottler to Applications Folder shortcut.

Go into the Applications folder and double-click Wine. You should see a wine glass icon in the upper right tray.

Double-click WineBottler in the Applications Folder. Select “Create Custom Prefixes”.

Download the most recent RecAll-PRO .EXE file and click the “select File…” button in the WineBottler window to select this file. (Under Winetricks an “allcodecs” package exists, but it doesn’t seem to refer to ACM codecs used by RecAll, so don’t click any packages.) Click the Install button.

In the pop-up dialog, select the Applications folder and change name to RecAll-PRO.

Run the RecAll-PRO installer and try out the program.

No Wave or MP3 compression codecs are available, but uncompressed Wave and all Vorbis/Speex compression works. (Perhaps some Wine/Mac guru can figure out how to make MP3 work?)  The next time you want to start the program, look in the Applications folder.


Related Posts:



Remember when voice recording was hard? Rare? A special event? Cost money? Required special hardware?


Full size ad at Vintage Ads.

More ads and hardware photos at Pinrepair.

Recording on YouTube.

(hat tip BoingBoing)

Related Posts:

Installing RecAll-PRO and RecAll on Ubuntu Linux

Although developed for Windows, Recall-PRO can be executed on the Linux operating system using the free Wine software compatibility layer. RecAll-PRO + Wine should also work on other flavors of Linux, but for the purposes of this article we focus on Ubuntu because of its current popularity and for the simple reason that some of our development machines uses this flavor. The procedure described has been tried on Natty Narwhal (version 11.04), Maverick Meerkat (10.10), and Oneiric Ocelot (11.10).

First, Wine must be installed. (We tend to use Synaptic Package Manager, but old-time Unix hands will open a Terminal Window and type “sudo apt-get” commands.) Start Synaptic Package Manager and enter “wine” in the search text field.

Select “wine1.3” (or later) and click the Apply button.

Download the latest RecAll-PRO install executable using your favorite web browser. In Linux, EXE file downloads do not have executable priveleges, so we have to change that now. Go to the Downloads section of your browser and <right-click> -> Open containing folder. (Note: Changing file executable permissions may not be necessary on recent versions of Ubuntu+Wine.)

Click the “Allow executing file as program” control.

In the folder, do not try to launch the RecAll-PRO EXE directly, but <right-click> -> Open with Wine Windows Program Loader.

This will start the installer to load RecAll-PRO and its help file on your computer.

Most features work as expected. MP3 recording does not work at present– perhaps later we will figure out how to access an Mpeg ACM codec under Wine. Vorbis and Speex compression work fine, because they do not rely on external codecs.

To start RecAll-PRO after installation:

Oneiric Ocelot (11.10) and the Unity desktop (and presumably all later versions): Use the Dash home button and start RecAll-PRO as with any native application. Coolness!

Natty Narwhal 11.04 and the Unity desktop: The first Unity iteration did not seem to have the best support for Wine applications. Add back the classic menu using the procedure here, and start using Classic menu-> Wine-> Programs-> Sagebrush-> RecAll-PRO.

Maverick Meerkat (10.10) (and possibly earlier versions) : Click applications menu-> Wine-> Programs-> Sagebrush-> RecAll-PRO.

Note: Earlier versions of RecAll-PRO had a problem displaying properly with Linux+Wine, as shown below. More recent versions (v1.9 and later) do not have the problem.


When Time Stamps Go Bad 2 reports Ohio electronic voting machines have privacy issues involving time stamps.

“Two Ohio activists have discovered that e-voting machines made by Election Systems and Software and used across the country produce time-stamped paper trails that permit the reconstruction of an election’s results–including allowing voter names to be matched to their actual votes.” — Declan McCullagh,

Our recording products have timecodes embedded in the audio files. Of course, our software is not involved in elections, but Sagebrush customers have imaginative applications, from security monitoring to documenting barking dogs to demonstrating a sleeping-partner snores. Given the news item shown here, can anyone think of a way that audio timecodes could cause an unplanned privacy problem?

Related post: When Time Stamps Go Bad

Recording in the News

Recording traffic stop results in felony charge. Link [via]

Passenger tapes interview with pilot and questions flight delay. Link [via]

Bike theft recording transformed into music video. Link [via]

Privatized downtown may prohibit photos/recording. Link [via]

Parents’ voice-activated recorder not admissible as evidence against bus driver. Link

R3c0rd phr0m H3ad53t C0nn3ct0r

Just for kicks, we wanted to see what one of our previous posts would look like if translated in “simple hax0r”, using a translator found on

Th3 0th3r day 50m30n3 call3d w1th an 1nt3r35t1ng pr0bl3m. H3 want3d t0 r3c0rd ht3 phar 51d3 uv a t3l3ph0n3 c0nv3r5at10n u51ng h1z lapt0p 50und d3v1c3, and h3 d1d n0t hav3 acc35z t0 a r3gular t3l3ph0n3 l1n3 t0 u53 0n3 uv ht3 t3l3ph0n3 r3c0rd1ng 1nt3rphac3z d35cr1b3d h3r3. H3 waz u51ng a V01P ph0n3 d3v1c3 that plugz 1nt0 a U5B p0rt, 50 n0 r3gular P0T5 l1n3 1z 1nv0lv3d. ht3 d3v1c3 pr0v1d3z a c0nn3ct0r ph0r a 5tandard t3l3ph0n3 h3ad53t, wh1ch c0mb1n3z m1cr0ph0n3 and 3arph0n35. 1n add1t10n, h3 want3d t0 5t1ll b3 abl3 t0 u53 h1z t3l3ph0n3 h3ad53t wh1l3 r3c0rd1ng, and k33p 3v3ryth1ng l0w-c05t, av01d1ng 50ld3r1ng 1ph p0551bl3.

1n th1z art1cl3 w3 d35cr1b3 a ph3w 51mpl3 d0-1t-y0ur53lph r3c0rd1ng 50lut10nz ph0r h3ad53t5. Th1z 1z al50 appl1cabl3 ph0r r3c0rd1ng phr0m a c3ll ph0n3 0r c0rdl35z ph0n3 w1th h3ad53t c0nn3ct0r, az w3ll az ht3 V01P ph0n3 d35cr1b3d ab0v3. Th353 50lut10nz m1ght N0T w0rk w1th land-l1n3 ph0n3z hav1ng h3ad53t c0nn3ct0r5, b3cau53 uv p0551bl3 gr0und l00p pr0bl3m5. W3 w1ll c0v3r th053 ph0n3z 1n a lat3r tut0r1al. Wh3n p0551bl3, w3 g1v3 Rad10 5hack part numb3r5– wh3r3 0mn1pr353nc3 c0untz m0r3 than qual1ty.

T3chn1cal Backgr0und

T3l3ph0n3 h3ad53tz m05t 0pht3n u53 2.5mm (3/32?) mal3 c0nn3ct0r5, 5mall3r than ht3 3.5mm (1/8?) aud10 c0nn3ct0rz ph0und 0n y0ur c0mput3r. H3r3 1z ht3 p1n-0ut uv ht3 h3ad53t plug:

C0mpar3 th1z t0 ht3 p1n-0ut uv y0ur l1n3-1nput c0nn3ct0r:

and m1cr0ph0n3 c0nn3ct0r:

V0ltag3 l3v3lz ph0r L1n3-1n R 0n ht3 0rd3r uv 1V0lt, and ht3 v0ltag3 l3v3l uv a m1cr0ph0n3 1z 0n ht3 0rd3r uv 10-100mV.

3xp3r1m3nt1ng w1th tw0 d1phph3r3nt c3ll ph0n3z and tw0 m0d3lz uv c0rdl35z ph0n3, w3 ph1nd ht3 5p3ak3r-ph0n3 1z d15abl3d wh3n ht3 h3ad53t plug 1z 1n53rt3d, pr3v3nt1ng uz phr0m m0n1t0r1ng ht3 ph0n3 call wh1l3 ht3 h3ad53t c0nn3ct0r 1z plugg3d, u51ng ht3 5am3 ph0n3.

51d3t0n3 r3ph3rz t0 ht3 phact that aud10 that w3 5p3ak 1n ht3 t3l3ph0n3 m1cr0ph0n3 1z add3d t0 what w3 h3ar at ht3 t3l3ph0n3 3arp13c3. Th1z 1z u53d az ph33dback 50 w3 d0n’t talk t0 l0ud and d15t0rt aud10, but m0r3 1mp0rtantly ph0r th1z appl1cat10n, th1z all0wz uz t0 r3c0rd ht3 aud10 at ht3 3arp13c3 and g3t b0th 51d3z 0ph ht3 c0nv3r5at10n.

(R3ad r35t uv art1cl3.)

Record from Headset Connector

The other day someone called with an interesting problem. He wanted to record the far side of a telephone conversation using his laptop sound device, and he did not have access to a regular telephone line to use one of the telephone recording interfaces described here. He was using a VOIP phone device that plugs into a USB port, so no regular POTS line is involved. The device provides a connector for a standard telephone headset, which combines microphone and earphones. In addition, he wanted to still be able to use his telephone headset while recording, and keep everything low-cost, avoiding soldering if possible.

In this article we describe a few simple do-it-yourself recording solutions for headsets. This is also applicable for recording from a cell phone or cordless phone with headset connector, as well as the VOIP phone described above. These solutions might NOT work with land-line phones having headset connectors, because of possible ground loop problems. We will cover those phones in a later tutorial. When possible, we give Radio Shack part numbers– where omnipresence counts more than quality.

Technical Background

Telephone headsets most often use 2.5mm (3/32″) male connectors, smaller than the 3.5mm (1/8″) audio connectors found on your computer. Here is the pin-out of the headset plug:

Compare this to the pin-out of your line-input connector:

and microphone connector:

Voltage levels for Line-In are on the order of 1Volt, and the voltage level of a microphone is on the order of 10-100mV.

Experimenting with two different cell phones and two models of cordless phone, we find the speaker-phone is disabled when the headset plug is inserted, preventing us from monitoring the phone call while the headset connector is plugged, using the same phone.

Sidetone refers to the fact that audio that we speak in the telephone microphone is added to what we hear at the telephone earpiece. This is used as feedback so we don’t talk to loud and distort audio, but more importantly for this application, this allows us to record the audio at the earpiece and get both sides of the conversation.

(Read rest of article.)

Convert WAV to MP3

I accidentally recorded a few audio files in .WAV file format instead of MP3. Is there an easy way to convert?

If the .WAV file was recorded using MPEG compression, simply rename the file to use an .MP3 file extension, and the file will play in the vast majority of MP3 player programs and MP3 devices.

Why does this work?

MP3 was designed to be a streaming format, and part of that design is to deliver the audio data in sections (called frames) and have a way to synch up to the next frame in case of loss of data, network problems, or starting playback in the middle of a stream. MP3 files often include non-audio data, called tags, at the beginning or end of the file, and player software is already designed to skip over tags to find the beginning of the first frame of audio.

Here is a conceptual schematic of one MP3 file with no tags:

[mp3 frame]
[mp3 frame]
[mp3 frame]

The WAV file format has a few extra bytes at the beginning of the file, but MP3 players will usually ignore bytes at the beginning of a file they don’t understand. Refer to this conceptual schematic of one WAV file:

‘RIFF’ [4 length bytes]
‘ fmt’ [4 length bytes]
‘data’ [4 length bytes]
[mp3 frame]
[mp3 frame]
[mp3 frame]

My WAV file does not use MPEG compression. Now what do I do?

Read rest of tutorial here.

Recording Office Meetings

If you use software to record office meetings, beware of unintended consequences.