In this paper, we propose a novel data hiding technique for G.711-coded speech based on the LSB substitution method. The novel feature of the proposed method is that a low-bitrate encoder, G.726 ADPCM, is used as a reference for deciding how many bits can be embedded in a sample. Experiments showed that the method outperformed the simple LSB substitution method and the selective embedding method proposed by Aoki. We achieved 4-kbit/s embedding with almost no subjective degradation of speech quality, and 10 kbit/s while maintaining good quality.

This paper proposes a bit-stream-arranged weighted modulation scheme to improve voice quality in low delay spread frequency selective fading channels. The proposed modulation scheme employs an input bit stream arrangement method that changes the bit stream order for significant bits so that they are not adjacent to each other over time; a mapping method that controls the amplitude of the modulation signals according to the importance of the bits; and modified differential encoding to prevent the error propagation from insignificant to significant bits. Computer simulations clarify that the proposed bit-stream-arranged weighted modulation scheme shows a S/N improvement of 8 dB in an 8-bit linear pulse code modulation (PCM) voice signal compared with the conventional non-weighted π/4-shift quadrature phase shift keying (QPSK) modulation scheme. The proposed scheme also shows 3. 5 dB improvement in a 4-bit adaptive differential pulse code modulation (ADPCM) voice signal. In this case, occurence of 'click noise' in recovered voice signal is halved. Although the proposed scheme increases the peak power of the modulated signals, the non-linearity of the power amplifier is not fatal.

This paper proposes a frequency diversity transmission scheme that obtains a frequency diversity gain and does not degrade spectrum efficiency; it utilizes multiple carrier frequencies alternately, not simultaneously. This scheme improves the bit error rate (BER) of significant information bits by sacrificing that of insignificant bits in fading channels. Simulation results show that the error floor of significant information bits is reduced to less than 1/5 while that of insignificant bits is doubled. They also show that the proposed scheme improves the received 4-bit ADPCM voice signal-to-noise ratio (SNR) by approximately 4 dB even when the frequency correlation is 0. 5.

Personal Handy Phone (PHP), the Japanese digital cordless telephone system is being developed. The 32kbits/s ADPCM (Adaptive Differential Pulse Code Modulation) codec has been standardized for PHP. This paper describes firstly, the advanced algorithms of a Voice Activity Detection (VAD) function that reduces power dissipation of a digital cordless telephone terminal, secondly, a comfort noise generator operates in conjunction with the VAD and finally, a transmission error control based on the use of the prediction coefficients generated in the ADPCM codec. These proposed algorithms function in the low signal-to-noise ratio (SNR) environment of personal radio communications. The quality of the reconstructed speech after the process is influenced by the VAD decision errors (false detection when no voice is present, or no detection when voice is present) , the similarity of the generated comfort noise to the actual background noise, and the transmission quality. The simulation results of the performance achieved by these algorithms are shown and required loading of the computation are also given.

Subjective quality tests have proven that embedded adaptive differential PCM (ADPCM), known to tolerate information loss through bit dropping, does not maintain sufficient speech quality when directly applied to asynchronous transfer mode (ATM) due to the fixed-length cell transmission scheme unique to ATM. We propose a coding and transmission scheme which enhances the performance by adjusting the embedded ADPCM coding rate according to input speech characteristics, thereby taking advantage of the ATM environment, where the transmission of variable rate sources is feasible. By varying the number of code bits of an embedded ADPCM coder from 6bits per sample, or 48kbps, for blocks of speech with a high prediction gain, to 2bits, or 16kbps, for silent blocks, a good compromise between coding bit rate and speech quality with gradual degradation due to information loss is achieved. The results of subjective evaluation tests showed the speech quality of the proposed scheme to be over 3.5 mean opinion score (MOS) on a scale of 1 to 5 at a cell loss rate of 10%. A prototype of the codec and the ATM cell assembly/disassembly functions were also fabricated using 3 conventional digital signal processors (DSPs) for real-time conversation tests.