Historique des commits

Auteur SHA1 Message Date
  Tristan Matthews be6df4d614 remove trailing whitespace il y a 9 ans
  jm 1f6cf519eb Moved the remaining of misc.h into arch.h il y a 17 ans
  jm bdcefb47d2 Put all the OS-dependent calls in os_support.h il y a 17 ans
  jm 9b70903cd3 Renamed the FFT in order to remove the symbol clash with Vorbis. il y a 20 ans
  conrad d8cc64be16 added guarded #include "config.h" throughout libspeex/ and src/ il y a 20 ans
  jm b9d3556440 fixed-point: converted QMF functions il y a 21 ans
  jm bf11f31650 fixed-point: rounding for shifts il y a 21 ans
  jm 19ad8233df fixed-point: merged floating-point and fixed-point functions (LPC and il y a 21 ans
  jm 99c7d0f18d fixed-point: LSP quantization work, also LSP's are now in the angle domain il y a 21 ans
  jm b1004ff2a9 fixed-point: converted user-visible functions to use "short" signals, il y a 21 ans
  jm 6c50b46a3f fixed-point: replace divisions by shifts... il y a 21 ans
  jm 3eae56c32d fixed-point: normalization function, some work on pitch closed-loop search il y a 21 ans
  jm 4b52842d69 fixed-point: pitch stuff il y a 21 ans
  jm 49d010a5ac fixed-point: conversion of the open-loop pitch analysis il y a 21 ans
  jm 0f3b00e4ba fixed-point: computation of rms values in fp il y a 21 ans
  jm 461f0eb357 fixed-point: removed some float's in innovation search il y a 21 ans
  jm c3d48cdf57 fixed-point: some innovation search details il y a 21 ans
  jm 9207e46ef1 fixed-point: most of the innovation search converted il y a 21 ans
  jm 306964514f fixed-point: some work on innovation quantization il y a 21 ans
  jm 8038eb0666 fixed-point: saturation for lsp_to_lpc, probably not the best solution il y a 21 ans
  jm 53f2631226 fixed-point: more conversion to spx_sig_t il y a 21 ans
  jm ad7cdd6531 cleaning up for 1.1 il y a 21 ans
  jm f66045e59e smallft.h doesn't need to be included from the .h files anymore il y a 21 ans
  jm e2ad040f15 the rest of the files for the Ephraim-Malah denoiser il y a 21 ans