Name recognition is an important application of automiatic speech recognition in embedded devices. Since embedded devices are used in diverse environments, noise robustness is very important. Moreover, unlike normal computer-based speech recognition applications, embedded speech recognition must deal with problems arising from limited resources. Facing these challenges, we have developed environment compensation and acoustic modeling techniques that improve robustness and accuracy of a speaker-independent name recognition system in hands-free conditions. These techniques are efficent to implement and are effective for performance improvement. On a name recognition task, we observed more than 53% word error rate reduction, compared to a baseline system. These improvements were obtained with minimal increase of resources.

This paper describes extensions of the 4 kb/s hybrid MELP/CELP coder, up to 6.4 kb/s and down to 2.4 kb/s. The baseline 4 kb/s coder uses three coding modes: MELP in strongly voiced speech frames, CELP with pitch prediction in weakly voiced frames, and CELP with stochastic excitation in unvoiced frames. To minimize switching artifacts between parametric MELP and waveform CELP coding, an alignment phase is encoded in MELP and zero-phase equalization is applied to the CELP target signal. The 6.4 kb/s extension uses the same three modes as the 4 kb/s coder, with improved MELP and CELP coders. The 2.4 kb/s extension uses only two modes: MELP for voiced frames and CELP synthesis with random excitation for unvoiced frames. The alignment phase is encoded in MELP frames for all bit rates so that time synchrony with input speech is always maintained. Alignment phase and zero-phase equalization enable smooth switching between coders at different bit rates. The hybrid MELP/CELP coding structure leads to coders that perform better at a given bit rate than MELP or CELP separately, and better than or equivalent to higher bit-rate ITU standards. Formal subjective tests show that for all-but-one tested conditions, the 6.4 kb/s hybrid coder is better than 8 kb/s G.729 and the 2.4 kb/s coder is equivalent to, or better than, 6.4 kb/s G.729 Annex D.

This paper presents an improved 4 kb/s hybrid MELP/CELP speech coder submitted as a candidate for ITU standardization. The coder uses three modes: a high-quality MELP coder for strongly voiced speech frames, an ACELP coder with pitch prediction for weakly voiced frames, and a stochastic CELP coder for unvoiced frames. We present recent enhancements to this coder, both to improve speech quality and to reduce coder complexity. Previous lTU Selection Testing results on an earlier version of this coder showed that it met nearly all requirements for toll-quality speech, more than any other candidate. Our internal testing shows that the current reduced-complexity fixed-point coder maintains this high performance.

Diversity schemes include information about packet n in future packets or send information about packet n via separate paths. If packet n is lost, it is reconstructed from information included in future packets or information received via separate paths. This paper presents CELP-based diversity schemes for voice over packet applications. The diversity schemes reduce the impact of packet losses while being efficient in terms of both bandwidth requirement and computational complexity. With our diversity schemes, transmission schemes that allocate bandwidth resources among diversity stages during congestion give significantly better performance than schemes that use no diversity during congestion, for the same bandwidth usage.

This paper describes new techniques for concealing frame erasures for CELP-based speech coders. Two main approaches were followed: interpolative, where both past and future information are used to reconstruct the missing data, and repetition-based, where no future information is required. Key features of the repetition-based approach include improved muting, pitch delay jittering, and LPC bandwidth expansion. The interpolative approach can be employed in voice over IP scenarios at no extra cost in terms of delay. Applied to the ITU-T G.729 ACELP 8 kb/s speech coding standard, both interpolation- and repetition-based techniques outperform standard concealment in informal listening tests.

With the advances in speech recognition and wireless communications, the possibilities for information access in the automobile have expanded significantly. We describe four system prototypes for (i) voice-dialing, (ii) Internet information retrieval-called InfoPhone, (iii) voice e-mail, and (iv) car navigation. These systems are designed primarily for hands-busy, eyes-busy conditions, use speaker-independent speech recognizers, and can be used with a restricted display or no display at all. The voice-dialing prototype incorporates our hands-free speech recognition engine that is very robust in noisy car environments (1% WER and 3% string error rate on the continuous digit recognition task at 0 db SNR). The InfoPhone, voice e-mail, and car navigation prototypes use a client-server architecture with the client designed to be resident on a phone or other hand-held device.

We have developed an adaptive multi-rate (AMR) speech coder designed to operate under the GSM digital cellular full rate (22.8 kb/s) and half rate (11.4 kb/s) channels and to maintain high quality in the presence of highly varying background noise and channel conditions. Within each total rate, several codec modes with different source/channel bit rate allocations are used. The speech coders in each codec mode are based on the CELP algorithm operating at rates ranging from 11.85 kb/s down to 5.15 kb/s, where the lowest rate coder is a source controlled multi-modal speech coder. The decoders monitor the channel quality at both ends of the wireless link using the soft values for the received bits and assist the base station in selecting the codec mode that is appropriate for a given channel condition. The coder was submitted to the GSM AMR standardization competition and met the qualification requirements in an independent formal MOS test.

A number of coding techniques have been reported to achieve near toll quality synthesized speech at bit-rates around 4 kb/s. These include variants of code excited linear prediction (CELP), sinusoidal transform coding (STC) and multi-band excitation (MBE). While CELP has been an effective technique for bit-rates above 6 kb/s, STC, MBE, waveform interpolation (WI) and mixed excitation linear prediction (MELP) models seem to be attractive at bit-rates below 3 kb/s. We present a system to encode speech with high quality using MELP, a technique previously demonstrated to be effective at bit-rates of 1.6-2.4 kb/s. We have enhanced the MELP model producing significantly higher speech quality at bit-rates above 2.4 kb/s. We describe the development and testing of a high quality 4 kb/s MELP coder.

It is well-known that an impulse-excited, all-pole filter is capable of representing many physical phenomena, including the oscillatory modes of percussion musical instruments like woodblocks, xylophones, or chimes. In contrast to the more common application of all-pole models to speech, however, practical problems arise in music synthesis due to the location of poles very close to the unit circle. The objective of this work was to develop algorithms to find excitation and filter parameters for synthesis of percussion instrument sounds using only an inexpensive all-pole filter chip (TI TSP50C1x). The paper describes analysis methods for dealing with pole locations near the unit circle, as well as a general method for modeling the transient attack characteristics of a particular sound while independently controlling the amplitudes of each oscillatory mode.

There is currently a great deal of interest in the development of speech coding algorithms capable of delivering toll quality at 4 kb/s and below. For synthesizing high quality speech, accurate representation of the voiced portions of speech is essential. For bit rates of 4 kb/s and below, conventional code excited linear prediction (CELP) may likely not provide the appropriate degree of periodicity. It has been shown that good quality low bit rate speech coding can be obtained by frequency domain techniques such as sinusoidal transform coding (STC), multi-band excitation (MBE), mixed excitation linear prediction (MELP), and multi-band LPC (MB-LPC) vocoders. In this paper, a speech coding algorithm based on an improved version of MB-LPC is presented. Main features of this algorithm include a multi-stage time/frequency pitch estimation and an improved mixed voicing representation. An efficient quantization scheme for the spectral amplitudes of the excitation, called formant weighted vector quantization, is also used. This improved coder, called mixed sinusoidally excited linear prediction (MSELP), yields an unquantized model with speech quality better than the 32 kb/s AD-PCM quality. Initial efforts towards a fully quantized 4 kb/s coder, although not yet successful in achieving the toll quality goal, have produced good output speech quality.

In general, a variable rate coder can obtain the same speech quality as a fixed rate coder, while reducing the average bit rate. We have developed a variable-rate multimodal speech coder with an average bit rate of 3 kb/s for a speech activity factor of 80% and quality comparable to the GSM full rate coder. The coder has four coding modes and uses a robust classification method involving the pitch gain, zero crossings, and a peakiness measure. Also the coder employs a novel gain-matched analysis-by-synthesis technique for very low rate coding of unvoiced frames and an improved noise-level-dependent postfilter. This paper describes the details of our algorithm and presents the results from subjective listening tests.

Remote retrieval of compressed voicemail data over a telephone line is one of several emerging applications of speech coding. Using a notebook computer equipped with a modem, a user can remotely access a networked desktop unit located at their office or home to retrieve various types of information such as email, FAX, electronic documents as well as voicemail. By compressing the speech data, we reduce the amount of time needed for the transfer of voice data over the telephone line. We have developed a high-quality variable-rate CELP (code-excited linear prediction) coder which can be used in such an application. The coder operates at an average rate of 3 kb/s assuming 80% speech activity and uses several new techniques including a mode decision mechanism based on the use of a peakiness measure and a novel excitation search method called gain-matched analysis-by-synthesis. The coder gives quality comparable to the Microsoft GSM 6.10 Audio Codec at 13 kb/s.

This paper presents a block-coded VFR strategy that removes the familiar "anchor frame" constraint in favor of a generalized delayed decision analysis. This approach is referred to as adaptive frame selection using dynamic programming (AFS/DP). The AFS/DP algorithm determines optimal breakpoint locations using a perceptually-based performance measure, accommodates fixed bit-rate operation, and operates with fixed delay and low complexity. The AFS/DP algorithm has been implemented in the context of the MELP vocoder at 2400 bits/sec, and has demonstrated comparable performance to block VFR methods, with lower coder delay.

In the digital cellular environment, acoustic background noise and channel bit errors (both bursty and random) may severely degrade the performance of speech coders such as the current European GSM full rate standard. This paper presents a CELP-based, enhanced full rate coder, which allocates 11.9 kb/s for speech and 10.9 kb/s for error protection for an aggregate channel rate of 22.8 kb/s. This coder provides toll quality speech which is significantly better than 16 kb/s ITU standard G.728 in clean conditions; also, it maintains robust performance over adverse conditions. The noteworthy features of our coder include: 10 msec frame size, leading to lower coder delay; improved error control coding; improved detection and masking of bad frames; low-complexity predictive tree-searched multi-stage vector quantization of the line spectrum pairs (designed using a noisy channel vector quantization algorithm); and low-complexity tree-searched multi-stage vector quantization of the fixed excitation.

This paper describes our enhanced mixed excitation linear prediction (MELP) speech coder which is a candidate for the new U.S. Federal Standard at 2.4 kbits/s. The new coder is based on the MELP model, and it uses a number of enhancements as well as efficient quantization algorithms to improve performance while maintaining a low bit rate. In addition, the coder has been optimized for performance in acoustic background noise and in channel errors, as well as for efficient real-time implementation. Listening tests confirm that the enhanced 2.4 kbit/s MELP coder performs as well as the higher bit rate 4.8 kbit/s FS1016 CELP standard.

We propose three new adaptive noise suppression algorithms for enhancing noise-corrupted speech: smoothed spectral subtraction (SSS), vector quantization of line spectral frequencies (VQ-LSF), and modified Wiener filtering (MWF). SSS is an improved version of the well-known spectra subtraction algorithm, while the other two methods are based on generalised Wiener filtering. We have compared these three algorithms with each other and with spectral subtraction on both simulated noise and actual car noise. All three proposed methods perform substantially better than spectral subtraction, primarily because of the absence of any musical noise artifacts in the processed speech. Listening tests showed preference for MWF and SSS over VQ-LSF. Also, MWF provides a much higher mean opinion score (MOS) than does spectral subtraction. Finally, VQ-LSF provides a relatively good spectral match to the clean speech, and may, therefore, be better suited for speech recognition.

A wavelet-transform-based CELP coder design is presented in this paper for high-quality speech coding at about 4.8 kbits/s. The coder quantizes the second residual using a wavelet transform approach instead of the stochastic-codebook-based vector quantization normally used in CELP coders, including the U.S. Federal Standard FS 1016 coder at 4.8 kbits/s. The wavelet coder improves the computational efficiency for encoding the second residual by requiring only 1.2 MIPS instead of 8.3 MIPS required by FS 1016. Subjective speech quality tests involving pairwise comparisons show that the wavelet coder was preferred 61% of the time over FS 1016.>

The authors present the result of their research on developing a hands-free voice communication system with a microphone array for use in an automobile environment. The goal of this research is to develop a speech acquisition and enhancement system so that a speech recognizer can reliably be used inside a noise automobile environment, for digital cellular phone application. Speech data have been collected using a microphone array and a digital audio tape (DAT) recorder inside a real car for several idling and driving conditions, and processed using delay-and-sum and adaptive beamforming algorithms. Performance criteria including signal-to-noise ratio and speech recognition error rate have been evaluated for the processed data. Detailed performance results presented show that the microphone array is superior to a single microphone.>

We present a method for word recognition with input speech transduced simultaneously by several sensors in high levels of broadband acoustic background noise. In prior work on single-input multisensor systems, limited success in machine recognition was achieved by linearly combining multiple sensor signals to yield a robust estimate of the speech signal in the presence of noise. In this paper, we demonstrate that improved recognition results are obtained by using all available sensor signals jointly as a vector, which preserves information from all sensors, as input to the decision process. We report on multisensor configurations using close-talking pressure-gradient microphones and accelerometers placed at the throat and nose of the speaker. The recognition error rates obtained by using the joint output vector are 45% lower than the error rates obtained with the best constituent sensor in the multisensor system; single-input multisensor systems, on the other hand, produce error rates that are about equal to the error rates obtained with the best constituent sensor.

In this paper, we report the results of isolated-word speech recognition tests performed with the Verbex 4000 recognizer on speech data collected 1) in 95 dB and 115 dB SPL broad-band acoustic noise typical in a fighter aircraft cockpit and 2) using several single sensors and two-sensor configurations that we reported in ICASSP-85. The two-sensor systems and the gradient microphones we tested produced about the same recognition performance in 95 dB noise. In 115 dB noise, the recognizer failed to train for all but the accelerometer, which produced nearly constant performance in both noise levels. Also, we demonstrate the feasibility of a new speech recognition methodology that uses a parallel system of multiple input signals transduced simultaneously using different sensors. In selected phonetic discrimination tests involving a feature-based approach, a parallel-input multisensor system reduced the discrimination errors to between one-half and one-twelfth of the number produced by a gradient microphone alone.

In this ongoing work, we have developed and tested configurations of multiple sensors for transducing speech with enhanced immunity to acoustic background noise. In this paper, we describe our work to improve upon our previously reported two-sensor configuration consisting of an accelerometer and a gradient microphone, and also present several additional multisensor configurations including one that uses a second-order gradient microphone for low frequencies and a first-order gradient microphone for high frequencies. Results from formal speech intelligibility and quality tests in simulated fighter aircraft cockpit noise show that each of the multisensor configurations tested outperforms the constituent individual sensors.

In this research, we developed and tested a variety of new objective measures for automatically evaluating pitch extractors for use in narrowband speech coders. For automatic computation of an accurate, reference pitch contour required by the objective measures, we modified an existing algorithm that uses the subglottal signal recorded simultaneously with the speech signal. To test and validate the objective measures, we conducted formal subjective speech quality tests, which involved five pitch extractors; a specially selected speech database of sentences that are likely to cause pitch and voicing errors; and two 2.4 kbit/s speech coders, the LPC coder and the harmonic deviations coder, each operating in clear or no noise and in Air Borne Command Post noise. Results obtained by correlating the objective scores with the mean subjective rating scores show that several of our objective measures produce consistently excellent correlation (0.9 - 0.995) for both coders and under both acoustic background conditions.

The aim of this work is to develop multisensor configurations of sensors for transducing speech in order to achieve enhanced immunity to acoustic background noise. We performed detailed measurements of the sound field in the vicinity of the mouth and neck during speech using pressure and pressure gradient microphones and an accelerometer. We investigated the properties of the measured signals from the various sensor types and positions through long-term and short-term spectral analyses and from articulation index scores computed assuming ambient noise typical of that in a fighter aircraft cockpit. From the results of this investigation, we developed a two-sensor configuration involving an accelerometer and a gradient microphone. Results from formal speech intelligibility and quality tests in simulated fighter aircraft cockpit noise show clearly that each of the two-sensor signals under test outperforms the signal from the gradient microphone alone and that the performance improvement generally increases with the noise level.

This paper describes the design of a baseband LPC coder that transmits speech over 9.6 kbit/s (kilobit/second) synchronous channels with random bit errors of up to 1 percent. Presented are the results of our investigation of a number of aspects of the baseband LPC coder with the goal of maximizing the quality of the transmitted speech. Important among these aspects are: bandwidth of the baseband, coding of the baseband residual, high-frequency regeneration, and error protection of important transmission parameters. The paper discusses these and other issues, presents the results of speech-quality tests conducted during the various stages of optimization, and describes the details of the optimized speech coder. This optimized speech coding algorithm has been implemented as a real-time full-duplex system on an array processor. Informal listening tests of the real-time coder have shown that the coder produces good speech quality in the absence of channel bit errors and introduces only a slight degradation in quality for channel bit error rates of up to 1 percent.

We review the variable frame rate (VFR) transmission methodology that we developed, implemented, and tested during the period 1973-1978 for efficiently transmitting LPC vocoder parameters extracted from the input speech at a fixed frame rate. In the VFR method, parameters are transmitted only when their values have changed sufficiently over the interval since their preceding transmission. We explored two distinct approaches to automatic implementation of the VFR method. The first approach bases the transmission decisions on comparisons of the parameter values of the present frame and the last transmitted frame. The second approach, which is based on a functional perceptual model of speech, compares the parameter values of all the frames that lie in the interval between the present frame and the last transmitted frame against a linear model of parameter variation over that interval. The application of VFR transmission to the design of narrow-band LPC speech coders with average bit rates of 2000-2400 bits/s is also considered. The transmission decisions are made separately for the three sets of LPC parameters, pitch, gain, and spectral parameters, using separate VFR schemes. A formal subjective spccch quality test of six selected LPC coders is described, and the results are presented and analyzed in detail. It is shown that a 2075 bit/s VFR coder produces speech quality equal to or better than that of a 5700 bit/s fixed frame rate coder.