This paper presents two FM receiver architectures and three digital signal processing algorithms for enhancing the output audio quality of FM broadcast receivers. The two receiver architectures differ only in the front-end processing for estimating the carrier-to-noise ratio and noise floors of the stereo audio signals. The shared back-end processing consists of three algorithms to suppress the noise in the audio signal, to detect and cancel noise pulses (static), and to conceal the degrading effects of fast fading, respectively. Together these FM enhancement techniques achieve about 20 to 35 dB improvements in SNR and stereo separation over a wide range of RF signal strength spanning nearly 30 dB. Perceptually, the audio quality improvement is large and obvious when the received FM signal is weak.

A class of packet loss concealment algorithms for speech coding is presented. It generates the replacement waveform for the lost frame by direct extrapolation of the past speech waveform, with or without look-ahead. The ITU-T G.722 Appendix III standard is based on it. When a future frame is unavailable (without look-ahead), the PLC algorithm gives significantly better speech quality than G.711 Appendix I - by about 0.2 PESQ for high packet loss rates. When a future frame is available (with look-ahead), the PLC algorithm uses the decoded speech waveform in the future frame to guide the pitch contour of waveform extrapolation during the lost frame such that the extrapolated waveform is phase-aligned with the decoded waveform after the packet loss. This technique further improved PESQ by another 0.2 for high packet loss rates.

This paper describes the BroadVoice® speech coding algorithm, which has been standardized as PacketCable™, SCTE®, and ANSI standards for VoIP cable telephony. The BroadVoice family of codecs includes a 16 kb/s BroadVoice 16 narrowband codec and a 32 kb/s BroadVoice32 wideband codec. BroadVoice is based on two-stage noise feedback coding with vector quantization of excitation and is optimized for low delay, low complexity, and high quality. It has an algorithmic delay of merely 5 ms and a complexity of only 12 and 17 MIPS for BroadVoice16 and BroadVoice32, respectively. The speech quality of BroadVoice16 is better than that of G.728, G.729, and 32 kb/s G.726. The speech quality of BroadVoice32 is better than that of 64 kb/s G.722.

This paper presents a candidate for the ITU-T G.722 packet loss concealment standard. The algorithm is based on waveform extrapolation in the speech domain. The strong backward adaptive nature of G.722 makes the state update during lost frames a challenge. The paper presents methods to update the G.722 subband decoder state memory during packet loss. Furthermore, novel techniques to facilitate smooth transition after packet loss are described. Formal subjective test results indicate that the algorithm far exceeds all requirements in the ITU-T terms of reference. Most techniques are applicable to G.726 as well, and some are easily extendable to other speech coders with memory.

This paper presents several novel codec structures for Noise Feedback Coding (NFC) incorporating both long-term and short-term noise spectral shaping, as well as long-term and short-term prediction. In addition, the paper generalizes the conventional scalar-quantization-based NFC to vector-quantization-based NFC, and it lays the foundation for the associated efficient VQ codebook search and closed-loop VQ codebook design. BroadVoice®16, a Packet Cable 1.5 mandatory narrowband speech codec standardized by Cable Labs®for Voice over Cable in North America, is based on one of such novel NFC codec structures.

This paper describes new efficient Vector Quantization (VQ) techniques that enable low complexity implementations of VQbased Noise Feedback Coding (NFC). These methods offer mathematical equivalence to higher complexity methods. Furthermore, the paper presents efficient codec structures for general noise shaping as used in BroadVoice®16 – a new SCTE ® (Society of Cable Telecommunications Engineers) and PacketCable ™ speech coding standard for Cable Telephony. Index Terms: speech coding, noise shaping, efficient VQ. 1.

This paper describes an embedded sinusoidal transform codec that is scalable not only in bit-rate, but also in sampling rate. In a representative implementation, the system produces an embedded bit-stream at 3.2 and 6.4 kbit/s for telephone-bandwidth speech, and scales up to 9.6 kbit/s for 16 kHz sampled wideband speech. The 3.2 kbit/s codec is a sinusoidal transform codec with synthetic phases. The 6.4 kbit/s codec adds resolution to the spectral envelope and transmits measured phases of the eight lowest harmonics. The 9.6 kbit/s codec adds information in the 4 to 8 kHz band to provide higher quality wideband speech.

This paper presents a high-fidelity speech and audio codec operating at a sampling rate of 32 kHz and a bit rate of 64 kbit/s. Designed primarily for real-time speech communication systems with high port densities, this MDCT-based transform codec has a very low coding delay (8 ms frame size) and low codec complexity (less than 10 MIPS on a 16-bit fixed-point DSP). The codec achieves essentially transparent quality for speech, and very close to transparent quality for music. A novel frame erasure concealment algorithm makes this codec robust to frame erasures for both speech and music. Another novel feature allows the decoder to decode the bit stream directly into a 16 kHz or 8 kHz sampled signal, without the need to decode a 32 kHz signal first and then down-sample it to the target sampling rate. Other novel features include some speed-memory trade-off techniques to reduce the computational complexity.

This paper presents a frame erasure concealment algorithm based on sinusoidal analysis-synthesis and its application to MDCT-based codecs. When a frame is lost, sinusoidal analysis of the previously decoded and buffered signal is performed. The analysis gives a set of sinusoids, which are used to synthesize the waveform corresponding to the lost frame. When applied to an MDCT-based codec, the algorithm exploits the MDCT windowing and overlap-add synthesis mechanism to improve waveform synthesis in the lost frame. At frame erasure rates up to 10%, this algorithm significantly improves perceptual quality of the decoder output.

This paper presents AT&T's candidate coder for the ITU-T's new wideband speech coding standard at 16, 24 and 32 kb/s. This coder achieves high speech quality with a low coder complexity. The basic idea of the coder is to perform closed-loop pitch prediction on perceptually weighted speech, and then quantize the prediction residual using perceptually based transform coding techniques. A first version of the coder based on DFT was thoroughly tested and submitted to the ITU-T in February 1996, and it was selected as one of two surviving candidates to advance to the next phase. A revised version based on MDCT was later submitted in October 1996. Both versions are described.

This paper presents a novel wideband speech coding algorithm called transform predictive coding (TPC). The main emphasis is on low complexity. TPC uses short-term and long-term prediction to remove the redundancy in speech. The prediction residual is quantized in the frequency domain based on a calculated noise masking threshold. In its simplest form, the TPC coder uses only open-loop quantization and therefore has a low complexity. A 16 kb/s full-duplex, open-loop TPC coder takes only 22% of the CPU load on a 150 MHz SGI Indy workstation and about 34% on a 90 MHz Pentium PC. The speech quality of TPC is almost transparent at 32 kb/s, very good at 24 kb/s, and acceptable at 16 kb/s. In the second half of the paper, we report our recent progress in using closed-loop quantization techniques to improve TPC output speech quality.

We present a 16 kb/s CELP coder with a complexity as low as 3 MIPS. The main thrust is to reduce the complexity as much as possible while maintaining toll-quality. This low-complexity CELP (LC-CELP) coder has the following features: (1) fast LPC quantization, (2) 3-tap pitch prediction with efficient open-loop pitch search and predictor tap quantization, (3) backward-adaptive excitation gain, and (4) a trained excitation codebook with a small vector dimension and a small codebook size. Most CELP coders require one full DSP or even two DSP chips to implement in real-time. In contrast, 3 to 6 full-duplex LC-CELP coders can fit into a single DSP chip, since each takes only around 3 MIPS to implement. This coder achieved slightly higher mean opinion stores (MOS) than the CCITT 32 kb/s ADPCM. It also exhibits good performance when tandemed with itself or transcoded with other coders.

We have improved G.728 output speech quality for frame erasure channels. Three cases are considered: (1) no change to G.728, (2) change only the G.728 decoder, and (3) change both the encoder and decoder. In case 1, we synthesize a bit-stream during erased frames so that the decoder decodes an excitation with low energy or with characteristics similar to the excitation of previous good frames. In case 2, the gain-scaled excitation and LPC coefficients are extrapolated, and vital operations of backward LPC and gain adaptations are continued. Case 3 adds spectral smoothing and increases bandwidth expansion for the LPC and gain predictors. These techniques are quite effective, as the speech quality degradation due to 1% frame erasures ranges from just slightly noticeable in case 1 to almost unnoticeable in case 3. For case 3, the output speech is still intelligible for frame erasure rates up to 10% or even 20%.

An adaptive postfiltering algorithm for enhancing the perceptual quality of coded speech is presented. The postfilter consists of a long-term postfilter section in cascade with a short-term postfilter section and includes spectral tilt compensation and automatic gain control. The long-term section emphasizes pitch harmonics and attenuates the spectral valleys between pitch harmonics. The short-term section, on the other hand, emphasizes speech formants and attenuates the spectral valleys between formants. Both filter sections have poles and zeros. Unlike earlier postfilters that often introduced a substantial amount of muffling to the output speech, our postfilter significantly reduces this effect by minimizing the spectral tilt in its frequency response. As a result, this postfilter achieves noticeable noise reduction while introducing only minimal distortion in speech. The complexity of the postfilter is quite low. Variations of this postfilter are now being used in several national and international speech coding standards. This paper presents for the first time a complete description of our original postfiltering algorithm and the underlying ideas that motivated its development.>

The CCITT is in the process of standardizing a 16 kb/s speech coder submitted by AT&T called low-delay code excited linear prediction (LD-CELP). In the first phase of CCITT testing, the coder met all performance requirements except for the tandeming condition. To improve the tandeming performance, the authors first tuned the perceptual weighting filter to optimize the coder's performance after three tandems. They then added an adaptive postfilter which was also tuned for three tandems. The excitation codebook was also reoptimized using a multiple-language IRS-weighted training database. With these changes, in the second phase of CCITT testing, the coder significantly exceeded the tandeming performance requirement and also met all other requirements. In fact, the speech quality of LD-CELP was equivalent to or better than that of 32 kb/s ADPCM for all conditions tested. It is anticipated that the version of LD-CELP will be formally ratified as a CCITT standard in 1992. The author describes the modifications made to improve the performance and discusses the CCITT test results.>

A low-delay code-excited linear prediction (LD-CELP) speech coder which is expected to be standardized in 1992 as a CCITT G Series Recommendation for universal applications of speech coding at 16 kb/s is presented. The coder achieves a one-way coding delay of less than 2 ms by making both the LPC predictor and the excitation gain backward-adaptive and by using a small excitation vector size of five samples. The official CCITT laboratory tests revealed that the speech quality of this 16 kb/s LD-CELP coder is either equivalent to or better than that of the CCITT G.721 standard 32-kb/s ADPCM coder for almost all conditions tested. A description of the LD-CELP algorithm, its implementation on the DSP32C for CCITT testing, and performance results from these tests are presented.>

The authors describe the algorithm modifications they have made in order to make the LD-CELP (low-delay code-excited linear prediction) algorithm suitable for 16 bit fixed-point implementation. They replaced T.P Barnwell's recursive windowing method (1981) by a novel hybrid which is partially recursive and partially nonrecursive. This method avoided the dynamic range problem and the double-precision arithmetic that would otherwise have been required. A fourth-order LPC filter was cascaded at the output of the original 50th-order LPC filter. This filter effectively reduced the spectral dynamic range of the output of the 50th-order LPC filter, therefore alleviating the ill-conditioning problem of the 50th-order LPC analysis. Although the algorithm of J. LeRoux and C. Gueguen (1979) was believed to be better suited for fixed-point arithmetic than Durbin's recursion (see L.R. Rabiner and R.W. Schafer, 1978), this was not necessarily true for the two-stage cascaded LPC filter used. Fixed-point simulation showed that the speech quality of the modified fixed-point LD-CELP algorithm was essentially the same as that of the original floating-point LD-CELP algorithm.>

A high-quality 16-kb/s speech coder which has a one-way coding delay of less than 2 ms is presented. The coder is basically a backward-adaptive version of the code-excited linear prediction (CELP) coder. The low coding delay is achieved by using backward-adaptive predictor and gain and by using an excitation vector size as small as five samples. The pitch predictor in conventional CELP coders is eliminated, and the linear predictive coding (LPC) predictor order is increased from 10 to 50. The excitation gain is updated by a tenth-order adaptive logarithmic gain predictor. This log-gain predictor and the LPC predictor are updated by performing LPC analysis on previous log-gain and coded speech, respectively. The excitation codebook is closed-loop optimized and the codebook index is Gray-coded to improve the robustness against channel errors. Formal subjective tests indicate that this coder produces high-quality speech comparable to that of the CCITT G.721 32-kb/s ADPCM standard.>

A real-time implementation of the low-delay code-excited linear prediction (LD-CELP) speech coder based on the AT&T DSP32C is described. The performance of the coder is described. Recursive windowing is used to calculate the autocorrelation coefficients by dividing the Levinson-Durbin recursion into smaller parts and spreading the computation over several speech vectors. The LD-CELP encoder uses about 90% of the processor time of an 80-ns DSP32C, while the decoder uses about 40%. Hence, the coder is implemented on two DSP32Cs. For a single encoding with clear or noisy channels, the speech quality produced by this coder is either equivalent to or better than the CCITT G.721 standard 32-kb/s ADPCM. However, for multiple asynchronous encodings the ADPCM coders gives slightly better performance. The LD-CELP coder passes dual-tone multifrequency tones and 300-, 1200-, and 2400-b/s modem signals.>

The authors explore the benefits of time-varying bit allocation to excitation and LPC (linear predictive coding) parameters for the case of codebook-excited LPC. The overall bit rate in the experiment was 4.8, 6.4, or 8.0 kb/s. In each case, permissible bit rates for the LPC component were 0, 24, 36, or 48 bits per frame, one of which was selected for each speech frame using a brute-force search maximum performance. Average SNR gains over conventional time-invariant methods were modest, on the order of 1 to 2 dB, but gains for certain speech segments were as high as 3 to 5 dB. Perceptually, gains due to variable bit allocation were most noticeable in the 6.4 kb/s system, especially with female speakers. However, even in this case, the benefits of flexible bit allocation were somewhat offset by distortions due to other inadequacies in the coding algorithm.>

A novel least-squares formulation of the vector linear prediction (VLP) problem is presented. Based on this formulation, we develop two new design methods for obtaining the optimal vector predictor for frame-adaptive prediction: the covariance method and the autocorrelation method, which bear the names of the corresponding methods in scalar LPC analysis. Our formulation reveals several previously unrecognized properties of the resulting normal equation. Simulation results for VLP of speech waveforms confirm that the two proposed methods indeed give higher prediction gain than previously developed methods.

An improved Vector APC (VAPC) speech coder at 4800 bps produces speech with very good communications quality while maintaining a complexity low enough to allow a real-time implementation with at most two commercially available DSP chips. The VAPC algorithm combines APC with vector quantization and incorporates analysis-by-synthesis, perceptual noise weighting, and adaptive postfiltering. A novel adaptive postfiltering technique helps to achieve an essentially inaudible level of coding noise. Real-time software has been developed for an implementation using the AT&T DSP32 floating-point processor chip. The overall complexity of the implemented VAPC system is about 3 million multiply-adds/second of computation and 6 kwords of memory.

The generalization of gain adaptation to vector quantization (VQ) is explored in this paper and a comprehensive examination of alternative techniques is presented. We introduce a class of adaptive vector quantizers that can dynamically adjust the "gain" or amplitude scale of code vectors according to the input signal level. The encoder uses a gain estimator to determine a suitable normalization of each input vector prior to VQ encoding. The normalized vectors have reduced dynamic range and can then be more efficiently coded. At the receiver, the VQ decoder output is multiplied by the estimated gain. Both forward and backward adaptation are considered and several different gain estimators are compared and evaluated. Gain-adaptive VQ can be used alone for "vector PCM" coding (i.e., direct waveform VQ) or as a building block in other vector coding schemes. The design algorithm for generating the appropriate gain-normalized VQ codebook is introduced. When applied to speech coding, gain-adaptive VQ achieves significant performance improvement over fixed VQ with a negligible increase in complexity.

We have developed a new speech coder which significantly enhances Adaptive Predictive Coding (APC) by using vector quantization. The coder, called Vector APC (VAPC), gives very good speech quality at 9.6 kb/s and reasonably good quality at 4.8 kb/s. In VAPC, redundancy is first removed by a long-delay predictor and then by a short-delay predictor; the prediction residual is then quantized by a gain-adaptive vector quantizer. In the receiver, decoded residual vectors are used to excite a synthesis filter to obtain the coded speech. The computations required by VAPC are only in the order of 2 to 4 million flops per second. Because of its low complexity and high speech quality, VAPC may offer a low-complexity alternative to Code-Excited Linear Prediction (CELP) at low bit rates.