Performance Evaluation of Speech Coders
Performance Evaluation of Speech Coders:
- There are two approaches to evaluating the performance of a speech coder in terms of its ability to preserve the signal quality
- Objective measures have the general nature of a signal-to-noise ratio and provide a quantitative value of how well the reconstructed speech approximates the original speech.
- Mean square error (MSE) distortion, frequency weighted MSE, and segmented SNR, articulation index are examples of objective measures.
- Objective measures are useful in initial design and simulation of coding systems .they do not necessarily give an indication of speech quality as perceived by the human ear
- The listener is the ultimate judge of the signal quality; subjective listening tests constitute an integral part of speech coder evaluation.
- Subjective listening tests are conducted by playing the sample to a number of listeners and asking them to judge the quality of the speech.
- The subjective tests are carried out in different environments to simulate real life conditions such as noisy, multiple speakers, etc.
- These tests provide results in terms of overall quality, listening effort, intelligibility, and naturalness. speech coders arehighly speaker dependent in that the quality varies with the age and gender ofthe speaker, the speed at which the speaker speaks and other factor
- The most popular ranking system is known as the mean opinion score or MOS ranking.
- This is a five point quality ranking scale with each point associated with a standardized descriptions: bad, poor, fair, good, excellent shown in table 2
- When digital speech-coded signal is transmitted from the mobile to the base station, and then demodulated into an analog signal which is then speech coded for retransmission as a digital signal over a landline or wireless link.is known as tandem signaling.
- As wireless systems, there will be a greater demand for mobile-to- mobile communications, and such links will involve at least two independent, noisy tandems
- In general, the MOS rating of a speech codec decreases with decreasing bit rate.Table3 gives the performance of some of the popular speech on the MOS scale.