Good Speech Quality Research Articles

Bone marrow transplantation in late infantile Batten disease and juvenile Batten disease.

Two patients, one each with LINCL and JNCL, underwent BMT in an attempt to alter the course of their disease. The LINCL patient received two transplants, one when asymptomatic, which was rapidly rejected, and a second BMT at age 3 years 9 months when the EEG had become abnormal. He has deteriorated in the following 2 years and now has only a few words, limited vision and feeding difficulties. Only two major seizures have occurred, but minor seizures are frequent. Although he is less severely affected than his sister at the same age, it is difficult to know if BMT has had any effect. The most significant difference is the relative mildness of seizure activity. The JNCL patient, transplanted at 8 years, when her visual problems had already started, is still in mainstream school, has good speech and good quality of life at age 9 1/2 years. No regression has been found, but EEG changes are now present and her vision has further deteriorated. The effect of BMT, if any, will not become apparent for several years in the more slowly progressive JNCL.

Secondary Pharyngeal Myotomy for Tracheoesophageal Speech

Pharyngeal myotomy in the post-laryngectomy patient has been described to reduce pharyngoesophageal spasm and facilitate vocal rehabilitation and tracheoesophageal speech. This report describes the complications and speech outcomes in 22 consecutive patients who underwent a secondary pharyngeal myotomy. The patients were all drawn from a hospital clinic dedicated to voice rehabilitation after laryngectomies. Their ages ranged from 54 to 81-years-old with a mean of 64. Two patients suffered surgical complications--both had pharyngocutaneous fistulas which resolved with conservative management. One patient had an incomplete myotomy and was treated with injection of boulinum toxin in the pharyngeal plexus. This treatment resulted in good vocalization. Seventeen of 22 patients in the study were long-term TEP users, and all of these were judged to have good speech quality. Our experience with myotomy for TEP speech demonstrates that many patients with pharyngoesophageal spasm can benefit from the procedure and successfully phonate. There were several complications related to the procedure, but these were successfully managed conservatively and without long-term sequela. Pharyngeal myotomy as an adjunct to TEP is useful and effective in selected patients.

An intelligent multimode voice communications system for indoor communications

A novel high-quality, low-complexity dual-rate 4.7 and 6.5 kbits/s algebraic code excited linear predictive codec is proposed for adaptive multi-mode speech communicators, which can drop their source rate and speech quality under network control in order to invoke a more error resilient modem amongst less favorable channel conditions. Source-matched binary Bose-Chaudhuri-Hocquenghem (BCH) codecs combined with unequal protection diversity- and pilot-assisted 16and 64-level quadrature amplitude modulation (16-QAM, 64-QAM) are employed in order to accommodate both the 4.7 and the 6.5 kbits/s coded speech bits at a signaling rate of 3.1 kBd. Assuming an excess bandwidth of 100%, in a bandwidth of 200 kHz 32 time slots can be created, which allows us to support in excess of 50 users, when employing packet reservation multiple access (PRMA). Good communications quality speech is delivered in an equivalent bandwidth of 4 kHz, if the channel signal-to-noise ratio (SNR) and signal-to-interference ratio (SIR) of the benign indoors cordless channel are in excess of about 15 and 25 dB for the lower and higher speech quality 16-QAM and 64-QAM systems, respectively, and the PRMA time-slots are sufficiently uninterfered due to using time-slot classification algorithms and due to the attenuation of partitioning walls and ceilings.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Variable bit‐rate CELP coding of speech with phonetic classification

AbstractA variable bit‐rate speech coder intended for digital cellular applications is described. A voice activity detection algorithm is used to distinguish active speech from background noise. Each frame of active speech is further classified to distinguish between three phonetic categories: voiced, unvoiced, and onset. Each input frame is assigned one of five bit rates according to voice activity and phonetic classification and coded using an analysis‐by‐synthesis algorithm tailored to the needs of the class that it belongs to. The resulting coder, called Variable Rate Phonetic Segmentation, produces good quality speech at an average bit‐rate below 3 kbit/s when operating with a voice activity factor of 0.5. Informal subjective quality assessment for speech in clean and noisy backgrounds indicates a performance that is comparable to the TIA standard QCELP algorithm while operating at a 25% to 40% lower average bit rate.

Low delay tree coding of speech at 8 kbit/s

A low delay tree coder for coding speech at 8 kbit/s that incorporates a new backward coefficient adaptation structure is presented. The coder has an algorithmic delay of 2.5 ms, produces good quality speech over ideal channels, and maintains robustness to bit errors for independent bit error rates up to and including 10/sup -2/. The tree coder consists of a randomly populated fractional rate tree, a weighted squared error distortion measure, the (M,l) tree search algorithm, an incremental path map symbol release rule, a long-term predictor, an all-pole short-term predictor, and the newly-obtained parameter adaptation algorithms; The sensitivity to channel errors is reduced by using the receiver excitation sequence for adaptation of the short-term predictor coefficients, but the ability to track rapid changes in the speech is retained by shaping the excitation sequence with an all-zero filter. The output speech of the tree coder is evaluated using segmental signal-to-noise ratios (SNRSEG), narrow band spectrograms, a paired comparison with pulse code modulation (PCM), and an equivalent noise paired comparison using modulated noise reference unit (MNRU).< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Analysis of the smoothed residual-driven algorithm for speech coders

The smoothed residual-driven (SRD) algorithm was devised recently to adapt the short-term predictor coefficients in low-delay speech coders. This algorithm uses a smoothed version of the prediction residual for coefficient adaptation to provide good-quality speech while maintaining robustness to channel errors. The SRD algorithm is further investigated here to understand the effect of the smoothing approximation on the algorithm tracking capability for tone, AR, and speech inputs in the presence of channel errors. The transversal structure SRD algorithm is studied here. The relationship between the minimum MSE values with the SRD and signal-driven (SD) algorithms is derived as a function of the predictor coefficients. Tracking analysis shows that the SRD algorithm can track the input if the step size /spl mu/ is sufficiently small. The LMS-SRD algorithm is stable and succeeds in tracking all inputs. The LMS-SRD algorithm is compared with other existing algorithms such as the SD, residual-driven (RD), and LMS-CCITT algorithms. >

Complexity reduction of CELP speech coders through the use of phase information

An efficient search procedure for CELP coders is described. Based on the phase spectra of the codevectors and the ideal excitation vector, a subset of the codebook is preselected. This subset is then exhaustively searched to find the best, perhaps optimum, codevector. Computational savings of the order of 30-40 are achieved while still maintaining good output speech quality.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Multiband linear predictive speech coding at very low bit rates

The authors describe the multiband linear predictive (MB-LPC) vocoder and its operation at 2.4 kb/s and 1.2 kb/s. The MB-LPC vocoder uses mixed excitation and exploits the advantages of both time and frequency domain speech coding techniques to produce natural sounding, good quality speech. Subjective performance of speech at 2.4 kb/s produced by the MB-LPC is very close to that for the 4.15 kb/s INMARSAT-M IMBE speech coder. Informal listening tests have shown that in most cases people could not tell the difference between the new 2.4 kb/s MB-LPC coder and the 4.15 kb/s INMARSAT-M IMBE coder.

Combined speech and channel coding for mobile radio communications

We present a combined speech and channel coding scheme for digital mobile communications. The speech coding algorithm is based on code-excited linear prediction (CELP) and it provides good communication quality speech at a rate of 4 kbps. For the channel code, both rate-compatible punctured convolutional (RCPC) codes and punctured Reed-Solomon (PRS) codes are considered. In the case of RCPC codes, soft decision decoding is considered, in addition to the simpler hard decision decoding. The modulation format chosen in our study is /spl pi//4-DQPSK with differential detection. Unequal error protection is used based on the bit error sensitivities of the different speech parameters. The performance of the combined speech and channel codec is studied under different mobile channel conditions, such as fade rates, signal-to-noise ratios, and interleaving delays. The results indicate that, with no interleaving delay and a large channel signal-to-noise ratio, PRS codes and RCPC codes with soft decision decoding have similar performance, given in terms of the segmental signal-to-noise ratio (SSNR) of the reconstructed speech. At low channel SNR, RCPC codes with soft decision decoding are significantly better than PRS codes. Informal listening tests were also conducted and the results indicate that the difference between the quality of the reconstructed speech in an error-free channel and that in a Rayleigh fading channel is imperceptible at channel SNRs larger than 20 dB for both the RCPC and the PRS-based codecs.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Transform representation of the spectra of acoustic speech segments with applications. II. Speech analysis, synthesis, and coding

For Part I see ibid., vol.1, no.2, p.180-95 (1993). In Part I of this paper, the authors introduced an approach to the representation of the speech spectral envelope which makes use of the Karhunen-Loeve (KL) transformation of acoustic subword segments. This signal-dependent representation captures, with a few KL vectors and transform coefficients, the perceptually and phonetically important structure of the spectral envelope. Here the authors apply this representation to the analysis, synthesis, and coding of speech. They propose simple quantization and coding strategies for the KL representation vectors as well as for the resulting transform coefficients. The resulting technique is a variable rate encoding scheme which achieves good speech quality at an average rate of 3.5 kb/s. >

Transmission of digitally encoded speech at 1.2 Kbaud for PCN

The transmission of digitally encoded speech at 1.2 Kbaud for mobile personal communication networks (PCN) is investigated both analytically and by computer simulations. Speech was encoded at 4.8 Kbit/s using a low complexity transform binary pulse excited LPC codec. A 64-level QAM modem was used having 3 sub channels that operated with different BERs. The sensitivity of the encoded speech bits to transmission errors was identified, and the bits classified into 3 groups. Each group was then individually coded by BCH codec of differing power. The output of the BCH coders were grey coded onto 3 QAM channels. By this arrangement the protection given to the speech bits was dependent on their vulnerability. The 6 bit QAM symbols were transmitted at 1.2 Kbaud over Rayleigh fading channels with pedestrian mobiles travelling upto 4 mile/h. For microcells using a propoagation frequency of 1.9 GHz and operating with channel SNRs in excess 26 dB we achieved good communication quality speech when second-order diversity and AGC were used.

On automatic estimation of articulatory parameters in a text-to-speech system

It is often conjectured that articulatory modelling of speech is desirable, if training data can be obtained. Articulatory parameters are likely to interpolate over long intervals, and reproduce many acoustic details from simple constraints governed by physical laws. For text-to-speech application, it may be easier to formulate rules in the articulatory domain. An analysis synthesis scheme for estimating the phoneme-level articulatory parameters to obtain best fits to natural speech, in the context of a text-to-speech (TTS) system, is presented. The working units of optimization are the parameters of an articulatory model (one vector per phoneme) and vectors of time and speed of transition for each parameter. The TTS system is used to initialize these parameters. Problems encountered in analysis synthesis are due to the following: (1) Automatic optimization methods work well only when the model is capable of matching the data closely. Otherwise, best solutions according to an objective error criteria are not necessarily sensible compromises according to perception. (2) Very different vocal tract shapes can produce speech spectra that are very similar (non-uniqueness). (3) Local minima can result from artifacts of the comparison strategy or due to the non-uniqueness in the spectrum-to-area transformation. Our solutions mainly consist of the following: (1) In each phoneme, adapt only those variables that have a large effect on the acoustic result (critical variables). (2) Use multiple error criteria sequentially to match large-scale spectral features in the first stage, and then, match the finer details using a high resolution criterion. It is demonstrated that good quality speech synthesis is possible through a phoneme-level control of an articulatory model. The synthesis is devoid of any artifacts, even in transitions, due to parameterization at the phoneme level. The quality of synthetic speech is comparable (at least in non-nasal voiced segments) to that of a good quality, frame-by-frame linear predictive coder. The models need to be embellished to reproduce other phonemes closely.

An efficient algorithm for articulatory synthesis.

One of the methods for efficient computation of articulatory synthesis is to discretize the area function into a small number of sections such as 10 or 12 resulting in a poor spatial resolution. The articulatory data have to be represented by 30 or more sections for a good spatial resolution requiring a large number of computations. However, good quality speech synthesis has been obtained using formant or LPC-based methods with filters of 10 or 12 coefficients for voiced sounds and 4 or 6 coefficients for unvoiced sounds. It is proposed that a computationally efficient method be used to obtain high quality synthesis using an articulatory model with a large number of sections (30 or more). It can be shown that when the adjacent sections are of equal area, the computations can be pruned without any loss of accuracy. For a given computational load, using an effective quantization rather than discretization of area function, one can achieve much lower spectral error in the transfer function. Trade-off relationships between spectral error and computational load as a function of quantization and discretization are reported.

Coarticulation rules in an articulatory model

Coarticulation rules are of primary importance for good speech quality in an articulation-based speech synthesis system. This is demonstrated by a simple perception experiment. Auditory feedback strategies combined with general phonetic knowledge about the relation between articulation and acoustics provide a reliable basis for the generation of such coarticulation rules.

Finite state CELP for variable rate speech coding

The performance of a variable rate code excited linear predictor system is investigated. The coding system is based on a finite state CELP (FSCELP) frame work. Each individual state is primarily identified with a LPC model order, LPC coefficients bit allocation, excitation code book population density and state encoding rate. Successive input speech vectors are encoded at a rate that depends on the current state of the FSCELP system and the input vector characteristics. The use of a finite state system involves implicit clustering of speech signals. The lower rate states are selected during highly correlated steady state speech segments when relatively few bits are required to obtain adequate fidelity. For speech signals with a strong glottal excitation, unvoiced signals and transient speech segments, a relatively greater quantisation accuracy is needed to obtain good fidelity and therefore higher rate states of the system are used. Further improvement is obtained by using gamma populated excitation codebooks, for those states that are mainly used to encode speech signals with a strong underlying glottal excitation pulses. Experiments focus on investigation of the varying encoding requirements of the excitation signal for low pass, voiced, unvoiced and transient speech signals. The parameters of the finite state CELP system are designed to match the encoding requirements of typical speech signals. The greater part of the coding gain is obtained from variable rate encoding of the excitation signal. Using a six-state FSCELP, good quality speech is obtained at an average, maximum and minimum bit rates of 4 kbit/s, 10 kbit/s and 2 kbit/s, respectively.

High quality multiband LPC coding of speech at 2.4 kbit/s

Until recently good quality digital speech transmission was not possible below about 9.6 kbit/s. With the development of code excited linear prediction (CELP) and multiband excitation (MBE) vocoders,2 high quality digital speech transmission became possible at bit rates as low as 4.8 kbit/s. Below 4.8 kbit/s however, vocoders such as LPC-103 and the channel vocoder4 which can only produce synthetic and unnatural speech are still in use. A new coding scheme termed the multiband LPC (MB-LPC) vocoder is proposed, which produces good and natural quality speech at 2.4 kbit/s. Subjective performance of speech at 2.4kbit/s produced by the CELP is very close to MBE coded speech at 4.8 kbit/s. Informal listening tests have shown that in most cases people could not tell the difference between the new 2.4kbit/s MB-LPC coder and the 4.8 kbit/s MBE vocoder.

Concatenated coding and interleaving for half-rate GSM digital mobile radio system

A concatenated Fire and convolutional coding scheme with burst error-trapping and eight-level soft-decision Viterbi decoding for the half-rate GSM digital mobile radio system, operating at a gross coded speed of 11.4 k bit/s is proposed. The concatenated coding scheme was designed in conjunction with a 6.7 kbit/s speech coder which offers good speech quality and robustness against noise disturbance. The simulated bit error probability peformance of the scheme is presented, and is compared with that of the full-rate GSM system.

Telecommunications Tutorial ‐ Part 2

As explained in Part 1, digitisation brings with it numerous benefits although digitised speech of good quality requires a bit rate of 64 kbps, which implies that a channel of 32 kHz will be required (see Part 1, page 61). The same speech in analogue form could be accommodated within a frequency band of 5 kHz, so digitisation requires more bandwidth.

食道音声の客観的評価法 母音の音質を対象として

A method of objectively evaluating the voice quality of esophageal speech is described in this paper. We used the vowel/a/ of 27 esophageal speakers as a voice sample.As judged from perceptual impressions, the voice samples were ranked by the paired comparison method. Acoustic parameters such as maximum phonation time, fluctuation of intensity and periodicity of vibration were examined. We compared the perceptual ranking and those three parameters by non-parametric statistical methods. There was a significant correlation between the last two parameters and the perceptual ranking. Low fluctuation (less than 10dB in 500msec from the voice on set) and periodic vibration were assumed to be a standard for the good voice quality of esophageal speech.

Adaptive dummy spectrum insertion method used in asynchronous speech scrambling

An adaptive dummy spectrum insertion method which is applied to the asynchronous analogue–digital–analogue (ADA) speech scrambler is presented. Related analysis and computer simulation results are given. Both analysis and simulation results indicate that the method greatly improves the security of the scrambler, and maintains good speech quality.