IRCAM to SPX Converter

IRCAM sound files originate from the Institut de Recherche et Coordination Acoustique/Musique — one of the world's foremost computer music laboratories, founded by composer Pierre Boulez in Paris. The format was created in the early 1980s to serve the research needs of IRCAM and has since been adopted by academic and artistic communities working at the intersection of science and sound. An IRCAM file begins with a 1024-byte header containing a magic number, sample rate, channel count, and an encoding type field that supports linear PCM (16/32-bit integer and 32-bit float), mu-law, and A-law variants. The header block also accommodates free-form annotation text, allowing researchers to embed experiment metadata directly in the audio file. Because the payload is uncompressed by default, recordings maintain full fidelity through successive analysis and resynthesis cycles — essential in psychoacoustic experimentation. Software such as Csound, libsndfile, and SoX reads and writes the format natively. Key advantages include a well-defined header that eliminates parsing ambiguity, support for floating-point samples essential in scientific DSP work, and deep roots in the computer music community ensuring continued tooling.

Speex is an open-source audio codec purpose-built for speech compression, developed by Jean-Marc Valin under the Xiph.Org Foundation. First released in October 2002, it targets voice-over-IP, conferencing, and any scenario where spoken word needs to travel efficiently over a network. SPX files wrap Speex-encoded audio inside an Ogg container, pairing the codec's speech optimization with Ogg's streaming capabilities. Three sampling rates are supported — narrowband at 8 kHz, wideband at 16 kHz, and ultra-wideband at 32 kHz — along with variable bitrate encoding that adapts in real time to speech complexity. A standout advantage is its patent-free, BSD-licensed nature, which allowed developers to embed it freely in both commercial and open-source products. Speex also bundles acoustic echo cancellation, noise suppression, and automatic gain control, features that rival codecs typically delegate to external libraries. Although its creators officially recommend Opus as a successor since 2012, Speex remains deployed in legacy VoIP systems, archived recordings, and embedded devices where its lightweight decoder footprint is still valued.