Transcoding of Speech
Overview of Outline:
Investigate transcoding between the two popular techniques of Waveform Interpolation (WI) and Mixed Excitation Linear Prediction (MELP). WI is an efficient algorithm for coding speech at low bit rates. High quality speech has been produced using a WI coder operating at 2.4 kpbs. MELP is the US military standard for speech coding at [url removed, login to view]. Although both WI and MELP fall within the parametric speech coding paradigm their structures are significantly different. This difference results in the translation from on bit stream to the other(transcoding) a complex task. Need to carried out the investigation below:
(1) Investigate transcoding by full
decoding and full re-encoding the speech signal. This is often called Tandem Transcoding. (see below (1))
(2) Investigate approaches that do not require decoding and re-encoding. E.g. Smart Mapping approaches. Come up with a proposed idea formal report and solution to ensure smart mapping.
(4)Compare quality of speech obtained using the two above approaches. The aim is to obtain speech of equivalent for both approaches.
1) Complete and fully-functional working program(s) in executable form as well as complete source code of all work done.
2)formal and detailed report of findings and results for both Comparison and Smart Mapping
3) Installation package that will install the software (in ready-to-run condition) on the platform(s) specified in this bid request.
4) Exclusive and complete copyrights to all work purchased. (No GPL, 3rd party components, etc. unless all copyright ramifications are explained AND AGREED TO by the buyer on the site).
WI runs in Visual C++
MELP runs in C, Currently using Cygwin to compile and run
Speeches can be heard by using Goldwave application.
my working OS is WIndows XP now.