About Voicegram

A Voicegram is a Public Voice Communication Audio Recording in Voice.

In the first Voicegram Recording Software implementation, the Free Software application Voice version 1.2.0, as a free purpose application, we can reproduce hearable sounds for human listening with time-space-frequency notation.

We use the principles in the processing of signals that are motivated by the processes involved in hearing.

A representation of audio signals where we have access to both time and frequency information is a well-motivated choice. The time-frequency domain is such a domain, and it is commonly deployed in audio processing. However, we want to add the extra capabilities of the Domain Name System information to annotate the full location representation with the unique time-space-frequency domain representation of the full audio signal in Voicegram Recording in Radio, the motivation behind Voice.

Specification

The Voicegram Recording 1.0 Specification is implemented in Voice 1.2.0 (Aamot, 2025) in ANSI C and available from http://folk.ntnu.no/olekaam/ntnu/bachelor/Aamot,2025.pdf

The Voicegram Recording Ogg Vorbis [.ogg] and XML file [.voice] is created in G_USER_DIRECTORY_MUSIC (usually $HOME/Music/GNOME.ogg and $HOME/Music/GNOME.voice on American English systems) with g_get_real_name():

<?xml version="1.0"?>

<voice version="1.2.0">

  <station name="Sondre Lerche" uri="http://www.sondrelerche.com/">

    <location lat="34.0223563" lon="-118.2873057">Los Angeles, California</location>

    <stream>http://api.perceptron.stream:8000/128.ogg</stream>

  </station>

</voice>

You’ll find the recorded Ogg Vorbis audio files along with the Voicegram Recording XML files in g_get_user_special_dir(G_USER_DIRECTORY_MUSIC) (usually $HOME/Music/) on GNOME 44 systems configured in the American English language.

The audio signals recorded with Voice version 1.2.0 have usually a sample rate of 44,100 Hz and are stored in the $HOME/Music/GNOME.ogg and $HOME/Music/GNOME.voice files for the label “GNOME”.

It is possible to configure multiple Voice listening streams with <stream></stream>. Your private Voice files is never shared with the public and can be kept private unless you manually share, upload or email the $HOME/Music/GNOME.ogg or $HOME/Music/GNOME.voice files with the public to a server such as GNOMEVOICE.ORG.

Implementation

The implementation of the Voicegram Recording 1.0 Specification (“as-is”) will be completed (“as-of”) on June 25th, 2024 in C as specified in The C programming language (Kernighan/Ritchie, 1978) after 6 months of work that began on February 15th, 2022 as Voice and the implementation of Voice was finished in Oslo, Norway by Ole Kristian Aamot on November 28nd, 2022.

References

Aamot, Ole Kristian: Public Voice Communication (NTNU, 2025)