FMJ-Software has updated Awave Audio to v10.3.

Changes:

• Added support for reading and writing Musepack (.MPC) compressed files (SV7 and SV8 types supported for reading, SV8 only for writing).
• Added support for reading and writing files with Shorten (.SHN) lossless compression.
• Several improvements to the normalization function:
  • Fixed a conformance problem with Replay Gain meta data calculations (if you have used an earlier version to calculate Replay Gain meta data for files, then it is recommended that you use a "tag editor" of your own choice to rescan them and calculate new Replay Gain data). This selection is now called "Replay Gain 'Standard'".
  • The new selection "Replay Gain 'ISO 226:2003'" adds a fairly accurate filter implementation based on the "75-phon equal loudness contour" from the "new" research standardized as ISO 226:2003. This is an alternative to the standard Replay Gain hearing filter (which is an approximation of the "80-dB F-weight curve", based on research as old as 1933!).
  • Also added new "Leq(RLB)" and "Leq(R2LB)" selections for loudness measurement (the latter is also know as ITU BS 1770). NB, these are currently only supported for 48KHz.
  In some informal tests that we've done, on music-only material Replay Gain with the ISO-based filter performed 14% better than Replay gain with the standard filter, and 18% better than Leq(R2LB). For speech material though, Leq(R2LB) outperformed both Replay Gain variants by 20-25%, although it should also be noted that the number of sound clips used for the speech tests were fairly small, so results may not hold up to closer statistical analysis.
• Several improvements to the "Dithering" option (called "Quantization" in previous versions):
  • Available dithering options are now: "None (round to nearest)", "White noise dither - Rectangular p.d.f. - +/-0.5 lsb", "White noise dither - Triangular p.d.f. - +/-1 lsb" (new in this version), and "Noise-shaped dither (IRSO 226:2003 HR) - Triangular p.d.f." (new in this version, better than old noise-shaped dither option that it replaces).
  • The selection box is now disabled when selecting a data format where no quantization has to be made (e.g. "Float 32-bit") or where the encoder accepts the data in floating point format and does its own quantization if necessary (e.g. .AAC, .MP2, .MP3, .MPC, .OGG).
  • The dithering selection is now stored in preset files.
• Added Unicode support for file names.
• Added Unicode support for text meta data (for file formats that support it). When saving to meta data formats (e.g. ID3v2 tags) that supports multiple character encodings, the program will now automatically select the most compact one (from Latin-1, UTF-8 or UTF-16) that can represent the text without losing any data.
• Preset files (.aap) are now stored in app-data folder (they were previously stored in the program folder, where Vista UAC didn't like them to be created).
• Added support for reading text meta data in .WAV files from a "cart chunk" used by some radio broadcasters.
• When reading a .CDA file (Windows place-holder for an Audio CD-track), then text meta data (artist, album, track name, track no, year) are now automatically retrieved from the freedb internet database.
• Added support for an "URL" (web-link) text meta data type.