A new sound coding strategy for suppressing noise in cochlear implants

In the n-of-m strategy, the signal is processed through m bandpass filters from which only the n maximum envelope amplitudes are selected for stimulation. While this maximum selection criterion, adopted in the advanced combination encoder strategy, works well in quiet, it can be problematic in noise as it is sensitive to the spectral composition of the input signal and does not account for situations in which the masker completely dominates the target. A new selection criterion is proposed based on the signal-to-noise ratio (SNR) of individual channels. The new criterion selects target-dominated (SNR > or = 0 dB) channels and discards masker-dominated (SNR<0 dB) channels. Experiment 1 assessed cochlear implant users' performance with the proposed strategy assuming that the channel SNRs are known. Results indicated that the proposed strategy can restore speech intelligibility to the level attained in quiet independent of the type of masker (babble or continuous noise) and SNR level (0-10 dB) used. Results from experiment 2 showed that a 25% error rate can be tolerated in channel selection without compromising speech intelligibility. Overall, the findings from the present study suggest that the SNR criterion is an effective selection criterion for n-of-m strategies with the potential of restoring speech intelligibility.     

Subspace algorithms for noise reduction in cochlear implants

A single-channel algorithm is proposed for noise reduction in cochlear implants. The proposed algorithm is based on subspace principles and projects the noisy speech vector onto "signal" and "noise" subspaces. An estimate of the clean signal is made by retaining only the components in the signal subspace. The performance of the subspace reduction algorithm is evaluated using 14 subjects wearing the Clarion device. Results indicated that the subspace algorithm produced significant improvements in sentence recognition scores compared to the subjects' daily strategy, at least in stationary noise. Further work is needed to extend the subspace algorithm to nonstationary noise environments.     

直接数据域局域联合混响抑制方法研究

针对非平稳条件下混响抑制问题,提出一种可利用待测数据统计信息的直接数据域局域联合(D³JDL)空时自适应方法。不同于级联方法,该方法没有独立运行直接数据域空时自适应和局域联合方法,仅在待测空时样本内用直接数据域空时自适应原理构造非期望数据矩阵,用于获取待测样本的统计信息,并作为局域联合方法的输入进行降维空时处理。由于D³JDL方法不存在待测样本与学习样本之间统计特性失配的问题,且较好地利用了统计信息,因此具备良好的非平稳环境适应能力。仿真和实验数据处理结果表明,D³JDL方法抗混响效果优于常规波束形成加匹配滤波处理和其它空时自适应处理方法。     

Intensity discrimination with cochlear implants

Intensity difference limens were measured for various frequencies and intensities of sinusoidal and pulsatile electrical stimulation in monkeys with electrodes implanted in the scala tympani, scala vestibuli, modiolus, or middle ear. Difference limens decreased, as a function of initial stimulus intensity, from values of 1.5-3 dB near threshold to as low as 0.5 dB near the upper limit of the dynamic range. If sensation level was held constant, difference limens decreased as a function of frequency up to about 500 Hz, and then remained constant. They were similar across a variety of electrode placements and separations if differences in threshold and dynamic range were taken into account. However, difference limens measured in severely damaged ears were slightly smaller than those in moderately damaged ears. The near miss to Weber's law, characteristic of acoustic difference limens, was not seen with electrical stimuli. Differences limens for electrical stimuli were roughly one-half those for acoustic stimuli; thus, part of the deficit in dynamic range for electrical stimulation compared with acoustic stimulation is countered by the smaller intensity differences limens for electrical stimuli.     

Place-pitch manipulations with cochlear implants

Pitch can be conveyed to cochlear implant listeners via both place of excitation and temporal cues. The transmission of place cues may be hampered by several factors, including limitations on the insertion depth and number of implanted electrodes, and the broad current spread produced by monopolar stimulation. The following series of experiments investigate several methods to partially overcome these limitations. Experiment 1 compares two recently published techniques that aim to activate more apical fibers than produced by monopolar or bipolar stimulation of the most apical contacts. The first technique (phantom stimulation) manipulates the current spread by simultaneously stimulating two electrodes with opposite-polarity pulses of different amplitudes. The second technique manipulates the neural spread of excitation by using asymmetric pulses and exploiting the polarity-sensitive properties of auditory nerve fibers. The two techniques yielded similar results and were shown to produce lower place-pitch percepts than stimulation of monopolar and bipolar symmetric pulses. Furthermore, combining these two techniques may be advantageous in a clinical setting. Experiment 2 proposes a method to create place pitches intermediate to those produced by physical electrodes by using charge-balanced asymmetric pulses in bipolar mode with different degrees of asymmetry.     

Forward masking in patients with cochlear implants

Forward masking was measured in 12 patients with cochlear implants. The amount of masking (in microamps) decreased linearly as a function of the logarithm of the signal delay from masker offset. Normalized forward-masking recovery functions for cochlear implants were similar to normalized functions of normal-hearing listeners, indicating that the mechanism of psychophysical forward masking is retrocochlear. These results indicate that the logarithm of acoustic amplitude should be mapped to microamps to produce normal forward masking in implanted patients. Despite the fact that the forward-masking recovery functions were similar across all patients, their performance with their speech processor varied widely.     

Localization cues with bilateral cochlear implants

Selected subjects with bilateral cochlear implants (CIs) showed excellent horizontal localization of wide-band sounds in previous studies. The current study investigated localization cues used by two bilateral CI subjects with outstanding localization ability. The first experiment studied localization for sounds of different spectral and temporal composition in the free field. Localization of wide-band noise was unaffected by envelope pulsation, suggesting that envelope-interaural time difference (ITD) cues contributed little. Low-pass noise was not localizable for one subject and localization depended on the cutoff frequency for the other which suggests that ITDs played only a limited role. High-pass noise with slow envelope changes could be localized, in line with contribution of interaural level differences (ILDs). In experiment 2, processors of one subject were raised above the head to void the head shadow. If they were spaced at ear distance, ITDs allowed discrimination of left from right for a pulsed wide-band noise. Good localization was observed with a head-sized cardboard inserted between processors, showing the reliance on ILDs. Experiment 3 investigated localization in virtual space with manipulated ILDs and ITDs. Localization shifted predominantly for offsets in ILDs, even for pulsed high-pass noise. This confirms that envelope ITDs contributed little and that localization with bilateral CIs was dominated by ILDs.     

Effects of reverberation and masking on speech intelligibility in cochlear implant simulations

Two experiments investigated the impact of reverberation and masking on speech understanding using cochlear implant (CI) simulations. Experiment 1 tested sentence recognition in quiet. Stimuli were processed with reverberation simulation (T=0.425, 0.266, 0.152, and 0.0 s) and then either processed with vocoding (6, 12, or 24 channels) or were subjected to no further processing. Reverberation alone had only a small impact on perception when as few as 12 channels of information were available. However, when the processing was limited to 6 channels, perception was extremely vulnerable to the effects of reverberation. In experiment 2, subjects listened to reverberated sentences, through 6- and 12-channel processors, in the presence of either speech-spectrum noise (SSN) or two-talker babble (TTB) at various target-to-masker ratios. The combined impact of reverberation and masking was profound, although there was no interaction between the two effects. This differs from results obtained in subjects listening to unprocessed speech where interactions between reverberation and masking have been shown to exist. A speech transmission index (STI) analysis indicated a reasonably good prediction of speech recognition performance. Unlike previous investigations, the SSN and TTB maskers produced equivalent results, raising questions about the role of informational masking in CI processed speech.     

人工耳蜗的环境声识别研究

对14位正常听力者开展了环境声的人工耳蜗仿真声识别实验,比较了两类声码器仿真(正弦载波和噪声载波)条件下的环境声识别效果差异,然后对9位讲普通话的成年人工耳蜗植入者开展了环境声识别实验。实验材料是从互联网上搜集,并经过12位正常听力者主观测试验证后,筛选出的67种环境声。结果显示,载波类型没有对67种环境声的平均识别效果产生显著影响,但是声学特征的差异会导致单个环境声的识别效果对载波类型有依赖。另外,人工耳蜗植入者的环境声识别效果较差,有待通过信号处理策略、神经接口和康复手段的改进而得到提高。本研究中开发的环境声材料可以用于评估人工耳蜗环境声识别效果。     

Speech discrimination in deaf subjects with cochlear implants

Electrical stimulation of the auditory nerve is being investigated as a way to provide information useful for speech communication in the profoundly deaf. Single-channel systems that tend to stimulate all fibers alike have had little success in achieving this goal. Multichannel systems that allow excitation of more complex temporal-spatial patterns of activity are now being introduced. Psychoacoustical experiments providing evidence that electrodes of a multichannel implant are able to separately excite distinct groups of neural elements are reviewed. New results using multiple electrodes and speech-like stimuli are presented. The synthetic stimuli were vowels (/a/, /i/, /u/) and consonant-vowel (CV) syllables (/ba/, /da/, /ga/, /ta/). Vowels and CV syllables were presented in an AXB discrimination task with different signal processing schemes and electrode configurations. A four-channel, frequency-selective system produced faultless discrimination scores for all stimuli and spontaneous recognition of the vowels while the scores for the single-channel system were generally much lower. Although understanding free running speech by the profoundly deaf does not seem imminent, the results presented indicate that the multichannel system tested shows more promise of approaching this goal than the single-channel scheme.     

Vowel spaces in Swedish adolescents with cochlear implants

This paper examines vowel production in Swedish adolescents with cochlear implants. Twelve adolescents with cochlear implants and 11 adolescents with normal hearing participated. Measurements were made of the first and second formants in all the nine long Swedish vowels. The values in hertz were bark-transformed, and two measures of the size of the vowel space were obtained. The first of them was the average Euclidean distance in the F1-F2 plane between the nine vowels and the mean F1 and F2 values of all the vowels. The second was the mean Euclidean distance in the F1-F2 plane between all the vowels. The results showed a significant difference for both vowel space measures between the two groups of adolescents. The cochlear implant users had a smaller space than the adolescents with normal hearing. In general, the size of the vowel space showed no correlations with measures of receptive and productive linguistic abilities. However, the results of an identification test showed that the listeners made more confusions of the vowels produced by speakers who had a small mean distance in the F1-F2 plane between all the vowels.     

Use of a sigmoidal-shaped function for noise attenuation in cochlear implants

A new noise reduction algorithm is proposed for cochlear implants that applies attenuation to the noisy envelopes inversely proportional to the estimated signal-to-noise ratio (SNR) in each channel. The performance of the proposed noise reduction algorithm is evaluated with nine Clarion CII cochlear implant patients using IEEE sentences embedded in multi-talker babble and speech-shaped noise at 0-10 dB SNR. Results indicate that the sigmoidal-shaped weighting function produces significant improvements to speech recognition compared to the subjects' daily strategy. Much of the success of the proposed noise reduction algorithm is attributed to the improved temporal envelope contrast.     

海底混响仿真研究

提出了一种新颖的海底混响仿真工程实现模型.它是基于单元散射的理论,通过模拟海底混响形成的物理过程实现仿真,具有仿真效果好,工程上易于实现的特点.根据该模型实现的多通道海底混响模拟器功能强、柔软性好,所产生的海底混响仿真信号无论是在形式上还是在统计特性方面与实测海底混响相比较都有很好的一致性.此外,该模拟器还可以作为目标模拟器或者相控/束控信号源使用.     

Acoustic cues to tonal contrasts in Mandarin: implications for cochlear implants

The present study systematically manipulated three acoustic cues--fundamental frequency (f0), amplitude envelope, and duration--to investigate their contributions to tonal contrasts in Mandarin. Simplified stimuli with all possible combinations of these three cues were presented for identification to eight normal-hearing listeners, all native speakers of Mandarin from Taiwan. The f0 information was conveyed either by an f0-controlled sawtooth carrier or a modulated noise so as to compare the performance achievable by a clear indication of voice f0 and what is possible with purely temporal coding of f0. Tone recognition performance with explicit f0 was much better than that with any combination of other acoustic cues (consistently greater than 90% correct compared to 33%-65%; chance is 25%). In the absence of explicit f0, the temporal coding of f0 and amplitude envelope both contributed somewhat to tone recognition, while duration had only a marginal effect. Performance based on these secondary cues varied greatly across listeners. These results explain the relatively poor perception of tone in cochlear implant users, given that cochlear implants currently provide only weak cues to f0, so that users must rely upon the purely temporal (and secondary) features for the perception of tone.     

Temporal pitch perception at high rates in cochlear implants

A recent study reported that a group of Med-El COMBI 40+CI (cochlear implant) users could, in a forced-choice task, detect changes in the rate of a pulse train for rates higher than the 300 pps "upper limit" commonly reported in the literature [Kong, Y.-Y., et al. (2009). J. Acoust. Soc. Am. 125, 1649-1657]. The present study further investigated the upper limit of temporal pitch in the same group of CI users on three tasks [pitch ranking, rate discrimination, and multidimensional scaling (MDS)]. The patterns of results were consistent across the three tasks and all subjects could follow rate changes above 300 pps. Two subjects showed exceptional ability to follow temporal pitch change up to about 900 pps. Results from the MDS study indicated that, for the two listeners tested, changes in pulse rate over the range of 500-840 pps were perceived along a perceptual dimension that was orthogonal to the place of excitation. Some subjects showed a temporal pitch reversal at rates beyond their upper limit of pitch and some showed a reversal within a small range of rates below the upper limit. These results are discussed in relation to the possible neural bases for temporal pitch processing at high rates.     

Better place-coding of the fundamental frequency in cochlear implants

In current cochlear implant systems, the fundamental frequency F0 of a complex sound is encoded by temporal fluctuations in the envelope of the electrical signals presented on the electrodes. In normal hearing, the lower harmonics of a complex sound are resolved, in contrast with a cochlear implant system. In the present study, it is investigated whether "place-coding" of the first harmonic improves the ability of an implantee to discriminate complex sounds with different fundamental frequencies. Therefore, a new filter bank was constructed, for which the first harmonic is always resolved in two adjacent filters, and the balance between both filter outputs is directly related to the frequency of the first harmonic. The new filter bank was compared with a filter bank that is typically used in clinical processors, both with and without the presence of temporal cues in the stimuli. Four users of the LAURA cochlear implant participated in a pitch discrimination task to determine detection thresholds for F0 differences. The results show that these thresholds decrease noticeably for the new filter bank, if no temporal cues are present in the stimuli. If temporal cues are included, the differences between the results for both filter banks become smaller, but a clear advantage is still observed for the new filter bank. This demonstrates the feasibility of using place-coding for the fundamental frequency.     

Development of a temporal fundamental frequency coding strategy for cochlear implants

A sound-coding strategy for users of cochlear implants, named enhanced-envelope-encoded tone (eTone), was developed to improve coding of fundamental frequency (F0) in the temporal envelopes of the electrical stimulus signals. It is based on the advanced combinational encoder (ACE) strategy and includes additional processing that explicitly applies F0 modulation to channel envelope signals that contain harmonics of prominent complex tones. Channels that contain only inharmonic signals retain envelopes normally produced by ACE. The strategy incorporates an F0 estimator to determine the frequency of modulation and a harmonic probability estimator to control the amount of modulation enhancement applied to each channel. The F0 estimator was designed to provide an accurate estimate of F0 with minimal processing lag and robustness to the effects of competing noise. Error rates for the F0 estimator and accuracy of the harmonic probability estimator were compared with previous approaches and outcomes demonstrated that the strategy operates effectively across a range of signals and conditions that are relevant to cochlear implant users.     

Coding of the fundamental frequency in continuous interleaved sampling processors for cochlear implants

In this study the perception of the fundamental frequency (F0) of periodic stimuli by cochlear implant users is investigated. A widely used speech processor is the Continuous Interleaved Sampling (CIS) processor, for which the fundamental frequency appears as temporal fluctuations in the envelopes at the output. Three experiments with four users of the LAURA (Registered trade mark of Philips Hearing Implants, now Cochlear Technology Centre Europe) cochlear implant were carried out to examine the influence of the modulation depth of these envelope fluctuations on pitch discrimination. In the first experiment, the subjects were asked to discriminate between two SAM (sinusoidally amplitude modulated) pulse trains on a single electrode channel differing in modulation frequency ( deltaf = 20%). As expected, the results showed a decrease in the performance for smaller modulation depths. Optimal performance was reached for modulation depths between 20% and 99%, depending on subject, electrode channel, and modulation frequency. In the second experiment, the smallest noticeable difference in F0 of synthetic vowels was measured for three algorithms that differed in the obtained modulation depth at the output: the default CIS strategy, the CIS strategy in which the F0 fluctuations in the envelope were removed (FLAT CIS), and a third CIS strategy, which was especially designed to control and increase the depth of these fluctuations (F0 CIS). In general, performance was poorest for the FLAT CIS strategy, where changes in F0 are only apparent as changes of the average amplitude in the channel outputs. This emphasizes the importance of temporal coding of F0 in the speech envelope for pitch perception. No significantly better results were obtained for the F0 CIS strategy compared to the default CIS strategy, although the latter results in envelope modulation depths at which sub-optimal scores were obtained in some cases of the first experiment. This indicates that less modulation is needed if all channels are stimulated with synchronous F0 fluctuations. This hypothesis is confirmed in a third experiment where subjects performed significantly better in a pitch discrimination task with SAM pulse trains, if three channels were stimulated concurrently, as opposed to only one.     

Frequency-to-electrode allocation and speech perception with cochlear implants

The hypothesis was investigated that selectively increasing the discrimination of low-frequency information (below 2600 Hz) by altering the frequency-to-electrode allocation would improve speech perception by cochlear implantees. Two experimental conditions were compared, both utilizing ten electrode positions selected based on maximal discrimination. A fixed frequency range (200-10513 Hz) was allocated either relatively evenly across the ten electrodes, or so that nine of the ten positions were allocated to the frequencies up to 2600 Hz. Two additional conditions utilizing all available electrode positions (15-18 electrodes) were assessed: one with each subject's usual frequency-to-electrode allocation; and the other using the same analysis filters as the other experimental conditions. Seven users of the Nucleus CI22 implant wore processors mapped with each experimental condition for 2-week periods away from the laboratory, followed by assessment of perception of words in quiet and sentences in noise. Performance with both ten-electrode maps was significantly poorer than with both full-electrode maps on at least one measure. Performance with the map allocating nine out of ten electrodes to low frequencies was equivalent to that with the full-electrode maps for vowel perception and sentences in noise, but was worse for consonant perception. Performance with the evenly allocated ten-electrode map was equivalent to that with the full-electrode maps for consonant perception, but worse for vowel perception and sentences in noise. Comparison of the two full-electrode maps showed that subjects could fully adapt to frequency shifts up to ratio changes of 1.3, given 2 weeks' experience. Future research is needed to investigate whether speech perception may be improved by the manipulation of frequency-to-electrode allocation in maps which have a full complement of electrodes in Nucleus implants.