CELT audio codec
CELT (Constrained Energy Lapped Transform) is an ultra-low delay audio codec designed for realtime transmission of high quality speech and audio. This is meant to close the gap between traditional speech codecs (such as Speex) and traditional audio codecs (such as Vorbis).
| Tags | Communications Telephony Internet Phone Conferencing multimedia Sound/Audio |
|---|---|
| Licenses | BSD Revised |
| Implementation | C |
Recent releases
- 06 Jul 2009 17:59
Changes: This release brings in many quality improvements, including better stereo coupling, better handling of transients, and better handling of highly tonal signals. In addition, packet loss robustness has been improved through the optional use of independent (intra) frames. CELT now supports a larger dynamic range, suitable for encoding 24-bit audio (float version only). There is also a very early VBR implementation.
- 18 Feb 2009 11:07
Changes: The pitch prediction was both improved and simplified. The second main change is a new bit allocation algorithm with better rounding and fine energy allocation. Other than that, there's a few bug fixes over 0.5.1.
- 18 Dec 2008 07:09
Changes: This release fixes a number of issues with 0.5.0. Several cases where the encoder could fail have been fixed, including one that could cause memory corruption (and possibly a segfault). In addition, the VQ search has been improved, which results in a small CPU usage reduction, while fixing some quality problems with low frequency tones. Other than that, there should be no significant quality changes.
- 11 Oct 2008 06:18
Changes: This release includes simplifications to the bit allocation code, and includes a new low-complexity mode that can be enabled (without breaking compatibility) on the encoder side. There are several other optimizations and bugfixes included. The API has been slightly modified, so code using CELT will have to be updated (minor).
- 27 Jul 2008 07:42
Changes: This release should greatly improve the handling of transients and reduce the amount of pre-echo, especially for larger frame sizes. There is now support for frame sizes up to 512 samples for those who don't have extreme delay requirements. The release also brings improved tuning for stereo and a variety of frame sizes and higher bit-rates. The API has been changed slightly to remove the overlap parameter, which is now automatically selected based on the frame size.
