SOU to HTK Converter

SOU is a raw audio format designation that functions as an alias for unsigned 8-bit PCM data (u8) in the SoX audio processing framework. Files with the .sou extension contain headerless, uncompressed audio samples stored as unsigned 8-bit integers — each byte represents a single amplitude value from 0 to 255, with 128 as the silence midpoint. Because there is no header, playback parameters such as sample rate and channel count must be specified externally. The default assumption is typically mono at 8000 Hz, though the data can represent any rate the recording hardware supported. The u8 encoding that SOU aliases is one of the simplest possible digital audio representations, predating structured audio containers like WAV and AIFF. Raw unsigned PCM was commonly produced by early sound cards and digitizers in the late 1980s and early 1990s, when storage constraints and limited processing power made headerless formats a practical choice. One advantage is absolute simplicity: SOU files can be read by any program capable of basic file I/O, with no parsing of container structures or metadata decoding required — useful for embedded systems, hardware diagnostics, and educational contexts where audio fundamentals are being explored. The format's minimal overhead also means that conversion to any modern container is lossless and instantaneous, since the raw PCM samples can be wrapped in a WAV or AIFF header without any transcoding.