文档库 最新最全的文档下载
当前位置:文档库 › Near-field adaptive beamformer for robust speech recognition

Near-field adaptive beamformer for robust speech recognition

Near-field adaptive beamformer for robust speech recognition
Near-field adaptive beamformer for robust speech recognition

Near-?eld Adaptive Beamformer for Robust

Speech Recognition

Iain A.McCowan Darren C.Moore S.Sridharan

Speech Research Laboratory,RCSAVT,School of EESE

Queensland University of Technology

GPO Box2434,Brisbane QLD4001,Australia

email:iain@https://www.wendangku.net/doc/a415780252.html,[dc.moore,s.sridharan]@https://www.wendangku.net/doc/a415780252.html,.au

This paper investigates a new microphone array processing technique speci?cally for the purpose of speech enhancement and recognition.The main objective of the

proposed technique is to improve the low frequency directivity of a conventional

adaptive beamformer,as low frequency performance is critical in speech processing

applications.The proposed technique,termed near-?eld adaptive beamforming

(NFAB),is implemented using the standard generalised sidelobe canceler(GSC)

system structure,where a near-?eld superdirective(NFSD)beamformer is used as

the?xed upper-path beamformer to improve the low frequency performance.In

addition,to minimise signal leakage into the adaptive noise canceling path for near-

?eld sources,a compensation unit is introduced prior to the blocking matrix.The

advantage of the technique is veri?ed by comparing the directivity patterns with

those of conventional?lter-sum,NFSD and GSC systems.In speech enhancement

and recognition experiments,the proposed technique out-performs the standard

techniques for a near-?eld source in adverse noise conditions.

Key Words:microphone array,beamforming,near-?eld,adaptive,superdirectivity,speech recog-nition

1.INTRODUCTION

Currently,much research is being undertaken to improve the robustness of speech recog-nition systems in real environments.This paper focuses on the use of a microphone array to enhance the noisy input speech signal prior to recognition.While the use of microphone arrays for speech recognition has been studied for some time by a number of researchers, a persistent problem has been the poor low frequency directivity of conventional beamfo-rming techniques with practical array dimensions.Low frequency performance is critical for speech processing applications,as signi?cant speech energy is located below1kHz. By explicitly maximising the array gain,superdirective beamforming techniques are able to achieve greater directivity than conventional techniques with closely spaced sensor arrays[1].This directivity generally comes at the expense of a controlled reduction in the white noise gain of the array.Recent work has demonstrated the suitability of superdirective beamforming for speech enhancement and recognition tasks[2,3].By

1

2I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

FIG.1.Generalised Sidelobe Canceler Structure

employing a spherical propagation model in its formulation,rather than assuming a far-

?eld model,near-?eld superdirectivity(NFSD)succeeds in achieving high directivity at low

frequencies for near-?eld speech sources in diffuse noise conditions[4].In previous work,

near-?eld superdirectivity has been shown to lead to good speech recognition performance

in high noise conditions for a near-?eld speaker[5].

Superdirective techniques are typically formulated assuming a diffuse noise?eld.While

this is a good approximation to many practical noise conditions,further noise reduction

would result from a more accurate model of the actual noise conditions during operation.

Adaptive array processing techniques continually update their parameters based on the

statistics of the measured input noise.The generalised sidelobe canceler(GSC)[6]presents

a structure that can be used to implement a variety of adaptive beamformers.A block

diagram of the basic GSC system is shown in Figure1.The GSC separates the adaptive

beamformer into two main processing paths-a standard?xed beamformer,,with constraints on the desired signal response,and an adaptive path,consisting of a blocking

matrix,,and a set of adaptive?lters,.As the desired signal has been constrained in the

upper path,the lower path?lters can be updated using an unconstrained adaptive algorithm,

such as the least-mean-square(LMS)algorithm.

While the theory of adaptive techniques promises greater signal enhancement,this is

not always the case in real situations.A common problem with the GSC system is

leakage of the desired signal through the blocking matrix,resulting in signal degradation

at the beamformer output.This is particularly problematic for broad-band signals,such

as speech,and especially for speech recognition applications where signal distortion is

critical.

In this paper we propose a system that is suited to speech enhancement in a practical

near-?eld situation,having both the good low frequency performance of near-?eld superdir-

ectivity,and the adaptability of a GSC system,while taking care to minimise the problem

of signal degradation for near-?eld sources.We begin by formulating a concise model

for near-?eld sound propagation in Section2.This model is then used in Section3to

develop the proposed near-?eld adaptive beamforming(NFAB)technique.To demonstrate

the bene?t of the technique over existing methods,an experimental evaluation assessing di-

rectivity patterns,speech enhancement performance,and speech recognition performance

is detailed in Sections4and5.

2.NEAR-FIELD SOUND PROPAGATION MODEL

NEAR-FIELD ADAPTIVE BEAMFORMER3

X

FIG.2.Near-?eld Propagation Model

In sensor array applications,a succinct means of characterising both the array geometry and the location of a signal source is via the propagation vector.The propagation vector concisely describes the theoretical propagation of the signal from its source to each sensor in the array.In this section,we develop an expression for the propagation vector of a sound source located in the near-?eld of a microphone array using a spherical propagation model.This expression is then used in the formulation of the proposed near-?eld adaptive beamformer in the following sections.

Many microphone array processing techniques assume a planar signal wavefront.This is reasonable for a far-?eld source,but when the desired source is close to the array a more accurate spherical wavefront model must be employed.For a microphone array of length ,a source is considered to be in the near-?eld if,where is the distance to the source,and is the wavelength.

We de?ne the reference microphone as the origin of a3-dimensional vector space,as shown in Figure2.The position vector for a source in direction,at distance from the reference microphone,is denoted,and is given by:

(1) The microphone position vectors,denoted as(),are similarly de?ned.The distance from the source to microphone is thus

(2) where is the Euclidean vector norm.

In such a model,the differences in distance to each sensor can be signi?cant for a near-?eld source,resulting in phase misalignment across sensors.The difference in propagation time to each microphone with respect to the reference microphone(),is given by

(3)

4I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

FIG.3.Near-?eld Adaptive Beamformer

where ms for sound.In addition,the wavefront amplitude decays at a rate proportional to the distance traveled.The resulting amplitude differences across sensors are negligible for far-?eld sources,but can be signi?cant in the near-?eld case.The microphone attenuation factors,with respect to the amplitude on the reference microphone, are given by

(4) Thus,if is the desired source at the reference microphone,the signal on the microphone is given by

(5) Consquently,we de?ne the near-?eld propagation vector for a source at distance and direction,as

(6)

3.NEAR-FIELD ADAPTIVE BEAMFORMING

The proposed system structure is shown in Figure3.The objective of the proposed technique is to add the bene?t of good low frequency directivity to a standard adaptive beamformer,as low frequency performance is critical in speech processing applications. The upper path consists of a?xed near-?eld superdirective beamformer,while the lower path contains a near-?eld compensation unit,a blocking matrix and an adaptive noise canceling ?lter.The principal components of the system are discussed in the following sections. Section3.1gives an explanation of the near-?eld superdirective beamformer.Section3.2 proposes the inclusion of a near-?eld compensation unit in the adaptive sidelobe canceling path,and examines its effect on reducing signal distortion at the output.Once this near-?eld compensation has been performed,a standard generalised sidelobe canceling blocking matrix and adaptive?lters can be applied to reduce the output noise power,as discussed in Section3.3.

NEAR-FIELD ADAPTIVE BEAMFORMER5

3.1.Near-?eld Superdirective Beamformer

Superdirective beamforming techniques are based upon the maximisation of the array gain,or directivity index.The array gain is de?ned as the ratio of output signal-to-noise ratio to input signal-to-noise ratio,and for the general case can be expressed in matrix notation as[1]:

(7) where is a column vector of channel gains,

(8)

is the complex conjugate transpose operator,and and are the cross-spectral density matrices of the signal and noise respectively.In practical speech processing applications the form of the signal and noise cross-spectral density matrices is generally unknown,and must be estimated,either from mathematical models(?xed beamformers) or from the statistics of the multi-channel inputs(adaptive beamformers).Superdirective beamformers are calculated based on assumed mathematical models for the and matrices.

When the desired signal is known to emanate from a single source at location, the signal cross-spectral matrix simpli?es to the propagation vector of the source,and the array gain can be expressed as:

(9) where is the propagation vector for the desired source,as de?ned in Equation6.

A diffuse(spherically isotropic)noise?eld is often a good approximation for many practical situations,particularly in reverberant closed spaces,such as in a car or an of?ce[7, 8].For diffuse noise,the noise cross-spectral density matrix can be formulated as:

(10) where is the propagation vector of a far-?eld noise source()in direction .

The superdirectivity problem is thus formulated as:

(11) By using a spherical propagation model to formulate the propagation vector,,the standard superdirective formulation can be optimised for a near-?eld source[9,4].As such,the only difference in the calculation of the standard and near-?eld superdirective channel?lters is the form of the propagation vector,.For a near-?eld source,the assumption of plane wave(far-?eld)propagation leads to errors in the array response to the desired signal due to curvature of the direct wavefront.A thorough discussion of the use of a near-?eld model for superdirective microphone arrays is given by Ryan and Goubran[9]. Cox[10]gives the general superdirective?lter solution subject to

6I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

1.linear constraints,(explained below);and

2.a constraint on the maximum white noise gain,,where is the desired white noise gain.

as

(12) where is a Lagrange multiplier that is iteratively adjusted to satisfy the white noise gain constraint.The white noise gain is the array gain for spatially white(incoherent)noise, that is,.A constraint on the white noise gain is necessary as an unconstrained superdirective solution will in fact result in signi?cant gain to any incoherent noise,partic-ularly at low frequencies.Cox[10]states that the technique of adding a small amount to each diagonal matrix element prior to inversion is in fact the optimum means of solving this problem.A study of the relationship between the multiplier and the desired white noise gain,shows that the white noise gain increases monotonically with increasing.One possible means of obtaining the desired value of is thus an iterative technique employing a binary search algorithm between a speci?ed minimum and maximum value for.The computational expense of the iterative procedure is not critical,as the beamformer?lters depend only on the source location and array geometry,and thus must only be calculated once for a given con?guration.

The constraint matrix,is of order,where there are linear constraints being applied,and the vector is a length-column vector of constraining values. The constraints generally include one specifying unity response for the desired signal, ,and where this is the sole constraint the above solution can by simpli?ed by substituting and,giving

(13) Once the optimal?lters have been calculated,the near-?eld superdirective beam-former output is calculated as

(14) where is the-channel input column vector

(15)

3.2.Near-?eld Compensation Unit

The?rst element in the adaptive path of standard GSC is the blocking matrix[6].Its purpose is to block the desired signal from the adaptive noise estimate.To ensure complete blocking,the desired signal must be both time aligned and have equal amplitudes across all channels.If this is the case,cancellation occurs if each row of the blocking matrix sums to zero,and all rows are linearly independent.

For a near-?eld desired source,to align the desired signal on all channels,a near-?eld compensation must?rst be applied to the input channels prior to blocking.To ensure full

NEAR-FIELD ADAPTIVE BEAMFORMER7

cancellation we need to compensate for both phase misalignment and amplitude scaling of the desired signal across sensors.We de?ne the diagonal matrix

diag(16)

where is the near-?eld propagation vector from Equation6.In this paper we de?ne the diagonal operator,diag,to produce a diagonal matrix from a vector parameter. Conversely,if invoked with a matrix parameter,it produces a row vector corresponding to the matrix diagonal.The near-?eld compensation can be applied as

(17)

Once this near-?eld compensation has been performed,a standard GSC blocking matrix can be employed to block the desired signal from the adaptive path.

The inclusion of this compensation unit is critical for a near-?eld desired signal.Without compensation for both phase and amplitude differences between sensors,blocking of the desired signal will not be ensured,leading to signal cancellation at the output.The near-?eld compensation effectively ensures that a true null exists in the beam-pattern of each blocking matrix row in the direction and distance corresponding to the desired source.To illustrate,Figure4shows the directivity pattern at kHz for the?rst row in the blocking matrix using the array shown in Figure5,with the desired source directly in front of the centre microphone at a distance of0.6m.The?gure shows the compensated response in the far-and near-?elds,as well as the uncompensated near-?eld response.It is clear that the uncompensated system will allow a high degree of signal leakage into the adaptive path as it blocks noise sources rather than the desired signal.

3.3.Blocking Matrix and Adaptive Noise Canceling Filter

The blocking matrix and adaptive noise canceling?lters are taken from the standard GSC technique[6].The order of the blocking matrix is,where there are

8I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

constraints applied in the?xed upper path beamformer.Generally only a unity constraint on the desired signal is speci?ed,and the standard Grif?ths-Jim blocking matrix is used:

.. ...

.

..

.

.. .

.. .

.. ...

.

..

.

(18)

The output of the blocking matrix is calculated as

(19) where is an-length column vector.De?ning the-length adaptive ?lter column vector as

(20) the output of the lower path is given as

(21) The NFAB output is then calculated from the upper and lower path outputs as

(22) and the adaptive?lters are updated using the standard unconstrained LMS algorithm

(23) where is the adaptation step size and denotes the current frame.

3.4.Summary of Technique

In summary,the proposed NFAB technique is characterised by the series of equations:

(24a)

(24b)

(24c)

(24d)

NEAR-FIELD ADAPTIVE BEAMFORMER9

noise

FIG.5.Experimental Con?guration

(24e) where all terms have been de?ned in the preceding discussion.

4.EXPERIMENTAL CONFIGURATION

For the experimental evaluation in this paper,we used the11element array shown in Figure5.The array consists of a9element broadside array,with an additional2 microphones situated directly behind the end microphones.The total array is40cm wide and15cm deep in the horizontal plane.The broadside microphones are arranged according to a standard broadband sub-array design,where different sub-arrays are used for different frequency ranges for the?xed upper path beamformer.The two end?re microphones are included for use by the near-?eld superdirective beamformer in the low frequency range. The four sub-arrays are thus

():microphones1-11;

():microphones1,2,5,8and9;

():microphones2,3,5,7and8;and

():microphones3-7.

The array was situated in a computer room,with different sound source locations,as shown in Figure5.The two sound sources were

1.the desired speaker situated60cm from the centre microphone,directly in front of the array;and

2.a localised noise source at an angle of124degrees and a distance of270cm from the array.

Impulse responses of the acoustic path between each source and microphone were mea-sured from multi-channel recordings made in the room with the array using the maximum length sequence technique detailed in Rife and Vanderkooy[11].As the impulse responses were calculated from real recordings made simultaneously across all input channels,they

10I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

TABLE1

Beamforming Techniques in Evaluation

technique description?lters

FS Conventional?lter-sum beamformer diag

NFSD Near-?eld superdirective beamformer

GSC GSC system with FS?xed upper path beamformer diag

NFAB Near-?eld adaptive beamformer

take into account the real acoustic properties of the room and the array.The multi-channel desired speech and localised noise microphone inputs were then generated by convolving the original single-channel speech and noise signals with these impulse responses.In addition,a real multi-channel background noise recording of normal operating conditions was made in the room with other workers present.This recording is referred to in the experiments as the ambient noise signal,and is approximately diffuse in nature.It consists mainly of computer noise,a variable level of background speech,and noise from an air-conditioning unit.The ambient noise effectively represents a diffuse noise?eld,while the localised noise represents a coherent noise source.In this paper,we specify the levels of the two different noise sources independently,as the signal to ambient-noise ratio(SANR)and signal to localised-noise ratio(SLNR).These values are calculated as the average segmental SNR from the speech and noise input,as measured at the centre microphone of the array. In this way,realistic multi-channel input signals can be simulated for speci?ed levels of ambient and localised noise.As well as facilitating the generation of different noise conditions,simulating the multi-channel inputs using the impulse response method is more practical than making real recordings for speech recognition experiments,as existing single channel speech corpora may be used.

5.EXPERIMENTAL RESULTS

This section presents the results of the experimental evaluation.The proposed NFAB technique is compared to a conventional?xed?lter-sum beamformer,a?xed near-?eld superdirective beamformer and a conventional GSC adaptive beamformer.These beam-formers are speci?ed in Table1.

The techniques are?rst assessed in terms of the directivity pattern in order to demon-strate the advantage of the proposed NFAB over conventional beamforming techniques, particularly at low frequencies.Following this,the techniques are evaluated for speech enhancement in terms of the improvement in signal to noise ratio,and the log area ra-tio.Finally,the techniques are compared in a hands-free speech recognition task in noisy conditions using the TIDIGITS database[12].

5.1.Directivity Analysis

As has been stated,the main objective of the proposed technique is to produce an adaptive beamformer that exhibits good low frequency performance for near-?eld speech sources.To assess the effectiveness of the proposed technique in achieving this objective, in this section we analyse the horizontal directivity pattern.The directivity of a?lter-sum beamformer is expressed in matrix notation as

(25)

(26)

5.1.1.Upper Path Directivity

First,we seek to demonstrate the directivity improvement that NFSD achieves at low frequencies compared to a conventional?lter-sum(FS)beamformer.For the FS beam-former,a common solution is to choose diag.This effectively ensures that the desired signal is aligned for phase and amplitude across sensors using a spherical propagation model.For NFSD,we use the?lter vector described in Section3.1. Figure6shows the near-?eld directivity pattern at300Hz for the FS and NFSD.From these?gures,it is clear that the NFSD technique results in greater directional discrimination at low frequencies compared to a conventional beamformer.At higher frequencies( kHz),conventional beamformers offer reasonable directivity,and so the FS and NFSD techniques give comparable performance.

5.1.2.Lower Path Directivity

Second,we wish to demonstrate the effect of the noise canceling path.The directiv-ity of the noise canceling?lters can be obtained by using the channel?lters

.The blocking matrix and adaptive?lters essentially implement a convention-al(non-superdirective)beamformer that adaptively focuses on the major sources of noise. To examine the directivity of the lower path?lters,the beamformer was run on an input speech signal with a white localised noise source(at the location shown in Figure5)added at an SLNR of0dB and a low level of ambient noise(SANR=20dB).The steady-state adaptive?lter vector,,was written to?le for both the proposed NFAB technique and the conventional GSC beamformer.The near-?eld directivity patterns of the lower path ?lters are plotted in Figures7and8for300Hz and5000Hz respectively.We see that the lower path adaptive?lters for both beamformers converge to similar solutions in terms of directivity,producing a main lobe in the direction of the coherent noise source(

12I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

5.1.3.Overall Beamformer Directivity

Finally,we examine the directivity pattern of the overall beamformer for the NFAB and conventional adaptive systems.The near-?eld directivity patterns at300Hz are shown in Figure9.We see that the directivity pattern of the NFAB system exhibits a true null in the direction and at the distance of the noise source,while the directivity of the conventional beamformer is too poor to signi?cantly attenuate the noise at this frequency.At frequencies above1kHz the directivity performance of both techniques is comparable.

5.1.4.Summary of Beamformer Directivity

In summary we see that,in terms of directivity,the proposed NFAB system:

outperforms the conventional FS system in terms of low frequency performance and the ability to attenuate coherent noise sources,

outperforms the NFSD system due to the ability to attenuate coherent noise sources, and

outperforms the conventional GSC system in terms of low frequency performance. In this way,we see that the proposed system succeeds in meeting the stated objectives and should therefore demonstrate improved performance in speech processing applications.

5.2.Speech Enhancement Analysis

The signal plots in Figure10give an indication of the level of enhancement achieved by the NFAB technique.For the desired speech signal,we used a segment of speech from the TIDIGITS database corresponding to the digit sequence one-nine-eight-six.Ambient noise was added at an SANR level of10dB,and a localised white noise signal was added at an SLNR level of0dB.The plots indicate that NFAB succeeds in reducing the noise level with negligible distortion to the desired signal.

To better measure the level of enhancement,objective speech measures were used to compare the different techniques.Two measures were used,these being the SNR improve-

14I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

FIG.10.Sample Enhanced Signal

ment and the log area ratio distortion measure.The SNR improvement is de?ned as the difference in SNR at the array output and input.As the true SNR cannot be measured, it is estimated as the average segmental signal-plus-noise to noise ratio.While the signal to noise ratio is a useful measure for assessing noise reduction,it does not necessarily give a good indication of how much distortion has been introduced to the desired speech signal.The log area ratio(LAR)measure of speech quality is more highly correlated with perceptual intelligibility in humans[13].The log area ratio measure for a frame of speech is calculated as

(27) where is the frame number,and and are the original and processed-order linear predictive coef?cients of the frame respectively.The overall log area ratio distortion measure for the signal is calculated as the average distortion over all input frames.

A set of experiments was conducted in which the localised white noise was replaced with a localised speech-like noise source taken from the NOISEX database[14].This is essentially a white noise signal that has been shaped with a speech-like spectral envelope, and thus represents a more realistic noise scenario than white noise.The signal to localised noise ratio(SLNR)was varied from20d

B to0dB,with the ambient noise present at a constant SANR level of10dB.The output signal to noise ratio improvement and log area ratios are given in Tables2and3for the different enhancement techniques1.The measures Sample sound?les are also available at https://www.wendangku.net/doc/a415780252.html,.au/speech/people/imccowan/

NEAR-FIELD ADAPTIVE BEAMFORMER 15

TABLE 2

Signal to Noise Ratio Improvement (SANR =10dB)

SLNR (dB)technique

05101520FS

0.50.40.30.10.1NFSD

1.4 1.6 1.10.50.2GSC

1.6 1.8 1.9

2.5

3.3NFAB 5.5 5.9 6.47.57.9

TABLE 3

Log Area Ratio :(SANR =10dB)

SLNR (dB)technique

05101520noisy input

3.6 3.3 2.9 2.6 2.5FS

3.1 2.7 2.5 2.4 2.3NFSD

2.8 2.3 2.1 2.0 1.9GSC

2.6 2.1 1.8 1.7 1.6NFAB 2.5 1.9 1.6 1.6 1.4

01

2

3

45678

9

SLNR (dB)S N R I m p r o v e m e n t (d B )(a) SNR Improvement 0152010500.511.522.533.54SLNR (dB)L o g A r e a R a t i o

(b) Log Area Ratio

01520105FIG.11.Speech Enhancement Measures :(a)SNR Improvement and (b)LAR

have been averaged over 10randomly chosen speech segments taken from different speakers in the TIDIGITS database.These results are plotted in Figure 11.

The SNR results show that the proposed NFAB gives considerably greater noise reduction compared to the FS,NFSD and GSC techniques,providing approximately 6-8dB of SNR improvement compared to the noisy input signal.Even with a relatively low level of localised noise (high SLNR),the NFAB technique offers signi?cantly greater noise reduction than these other methods.In addition,the proposed technique gives less distortion than the other techniques,as measured by the LAR.As would be expected,the ?xed NFSD technique gives slightly less distortion than the adaptive GSC technique.

16I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

TABLE 4

Speech Recognition Results :Word Recognition Rates

SLNR (dB)technique

1050-5noisy input

86.865.923.213.1FS

89.281.762.936.4NFSD

97.793.277.245.4GSC

88.883.873.856.8NFAB 98.296.791.176.7

10

20

30

40

50607080

90

100

105

0-5SLNR (dB)W o r d R e c o g n i t i o n R a t e (%)

FIG.12.Speech Recognition Results :Word Recognition Rates

From these results we see that,in a high level of diffuse and coherent noise with a near-?eld desired speech source,the proposed NFAB technique succeeds in signi?cantly reducing the noise level,while minimising the distortion to the speech signal.

5.3.Speech Recognition Analysis

The same noise scenario was used for experiments in robust speech recognition.The training and test data for the experiments was taken from the male adult portion of the TIDIGITS database.Tied-state triphone hidden Markov models and standard MFCC parameterisation with energy,delta and acceleration coef?cients were used.The models were trained with the clean input to the centre microphone,and then re?ned using MAP adaptation to better match the noisy environment.The noise segments used in the adaptation process were taken from a separate recording made in the room.The recognition results are given as percentage word recognition rates in Table 4and shown graphically in Figure 12.The results clearly show that NFAB gives excellent robustness to adverse noise conditions in a near-?eld speech recognition application.The results at low noise levels show that the baseline recognition system is already quite robust to noise,due to the use of MAP adaptation.At more realistic noise levels,however,unenhanced performance is clearly

NEAR-FIELD ADAPTIVE BEAMFORMER17 unsatisfactory.For example,at an SLNR of0dB and SANR of10dB,the word error rate for the unprocessed input is76.8%.While standard GSC and NFSD are able to reduce this to26.2%and22.8%respectively,the proposed NFAB technique succeeds in reducing the error rate to8.9%.As would be expected,the?gure shows that NFAB offers similar performance to NFSD when the noise is approximately diffuse(high SLNR),and demonstrates improved ability to attenuate any coherent noise sources(low SLNR)due to the GSC-style adaptive noise canceling path.The recognition performance of NFAB is seen to be similar to NFSD at low levels of coherent noise(high SLNR),and degrades at a rate comparable to GSC with increasing levels of coherent noise.

It is apparent from these results that NFAB is an enhancement technique that is well suited to speech recognition.The experimental results for both speech enhancement and recognition demonstrate that for an adaptive beamformer to be applicable in speech processing applications,it should exhibit good directivity at low frequencies,and take care to minimise any signal degradation.

6.CONCLUSIONS

A new microphone array processing technique designed speci?cally for near-?eld speech processing applications has been proposed,termed near-?eld adaptive beamforming(N-FAB).The technique incorporates a?xed near-?eld superdirective beamformer into a GSC-style adaptive beamforming structure,and as such exhibits the bene?ts of good low frequency performance and the ability to adaptively attenuate coherent noise signals.Dis-tortion due to the adaptive noise canceling path is minimised by the introduction of a near-?eld compensation unit.

Two major problems with common conventional microphone array techniques are their poor low frequency performance and the introduction of signal distortion in adaptive techniques.By taking care to address both these issues,the proposed NFAB technique succeeds in signi?cantly outperforming conventional beamforming techniques in terms of objective speech quality measures and speech recognition results in both diffuse and coherent noise.

Speech enhancement results indicate that NFAB succeeds in signi?cantly reducing the output noise power,while also minimising the distortion to the desired signal.These characteristics make it ideal as an enhancement technique for robust speech recognition. In a high noise con?guration,with a signal to localised noise ratio of0dB,and a signal to ambient noise ratio of10dB,the proposed technique succeeds in increasing the recognition rate from23.2%to91.1%.For the same con?guration,near-?eld superdirectivity and conventional GSC only achieve77.2%and73.8%respectively.

In summary,near-?eld adaptive beamforming has been shown to be a speech enhance-ment technique that produces a high quality,highly intelligible signal for applications requiring hands-free speech acquisition where the desired speaker is in the array’s near-?eld.

ACKNOWLEDGEMENTS

The near-?eld superdirective technique was developed at France Telecom R&D[4].The initial application of these techniques in a speech recognition application was researched by the author during a six month research period at France Telecom R&D in1999[5], under the supervision of Yannick Mahieux.The authors speci?cally wish to acknowledge Claude Marro for his helpful and insightful review of this paper.

18I.A.MCCOWAN,D.C.MOORE,S.SRIDHARAN

REFERENCES

1.H.Cox,R.Zeskind,and T.Kooij.Practical supergain.IEEE Transactions on Acoustics,Speech and Signal

Processing,ASSP-34:393–398,June1986.

2.J.Bitzer,K.U.Simmer,and K.Kammeyer.Multi-microphone noise reduction techniques for hands-free

speech recognition-a comparative study.In Robust Methods for Speech Recognition in Adverse Conditions (ROBUST-99),pages171–174,Tampere,Finland,May1999.

3.M.Doerbecker.Speech enhancement using small microphone arrays with optimized directivity.In Proc.Int.

Workshop on Acoustic Echo and Noise Control,pages100–103,September1997.

4.W.T¨a ger.Near?eld superdirectivity(NFSD).In Proceedings of ICASSP’98,pages2045–2048,1998.

5.I.McCowan,C.Marro,and L.Mauuary.Robust speech recognition using near-?eld superdirective beamfo-

rming with post-?ltering.In Proceedings of ICASSP2000,volume3,pages1723–1726,2000.

6.L.Grif?ths and C.Jim.An alternative approach to linearly constrained adaptive beamforming.IEEE Trans.

on Antennas and Propagation,30(1):27–34,January1982.

7.J.Meyer and K.Uwe Simmer.Multi-channel speech enhancment in a car environment using wiener?ltering

and spectral subtraction.In Proceedings of ICASSP97,volume2,pages1167–1170,1997.

8.J.Bitzer,K.Uwe Simmer,and K.Kammeyer.Theoretical noise reduction limits of the generalized sidelobe

canceller(gsc)for speech enhancement.In Proceedings of ICASSP99,volume5,pages2965–2968,1999. 9.J.G.Ryan and R.A.Goubran.Near-?eld beamforming for microphone arrays.In Proceedings of ICASSP

97,pages363–366,1997.

10.H.Cox,R.Zeskind,and M.Owen.Robust adaptive beamforming.IEEE Transactions on Acoustics,Speech

and Signal Processing,35(10):1365–1376,October1987.

11.D.Rife and J.Vanderkooy.Transfer-function measurement with maximum-length sequences.Journal of the

Audio Engineering Society,37:419–444,June1989.

12.Texas Instruments and NIST.Studio quality speaker-independent connected-digit corpus(TIDIGITS).CD-

ROM,February1991.NIST Speech Discs4-1,4-2and4-3.

13.S.R.Quackenbush,T.P.Barnwell,and M.A.Clements.Objective Measures of Speech Quality.Prentice-Hall,

NJ,1988.

14.Defence Research Agency Speech Research Unit.NOISEX-92.CD-ROM,June1992.

BIOGRAPHIES

Iain A.McCowan received the B.Eng(Hons)and https://www.wendangku.net/doc/a415780252.html,Tech degrees from the Queens-land University of Technology,Brisbane,in1996.In February1998he joined the Research Concentration in Speech,Audio and Video Technology at the Queensland University of Technology where he is currently completing his Ph.D.His main research interests are in the?elds of robust speech recognition and speech enhancement using microphone arrays. Mr McCowan is a student member of the Institute of Electrical and Electronic Engineers.

Darren C.Moore received the B.Eng(Hons)and https://www.wendangku.net/doc/a415780252.html,Tech degrees from the Queens-land University of Technology,Brisbane,in1997.In February1998he joined the Research Concentration in Speech,Audio and Video Technology at the Queensland University of Technology,where he is currently completing a M.Eng degree.His professional interests lie in the?eld of speech enhancement using microphone arrays and in the implementation of real-time DSP solutions.

Dr.S.Sridharan obtained his B.Sc(Electrical Engineering)and M.Sc(Communication Engineering)from the University of Manchester Institute of Science and Technology, United Kingdom and Ph.D(Signal Processing)from the University of New South Wales, Australia.Dr.Sridharan is Senior Member of the IEEE,USA and a Corporate Member of IEE,United Kingdom and IEAust of Australia.He is currently a Professor in the School of Electrical and Electronic Systems Engineering of the Queensland University of Technology (QUT)and is also the Head of the Research Concentration in Speech,Audio and Video Technology at QUT.

初中英语介词用法总结

初中英语介词用法总结 介词(preposition):也叫前置词。在英语里,它的搭配能力最强。但不能单独做句子成分需要和名词或代词(或相当于名词的其他词类、短语及从句)构成介词短语,才能在句中充当成分。 介词是一种虚词,不能独立充当句子成分,需与动词、形容词和名词搭配,才能在句子中充当成分。介词是用于名词或代词之前,表示词与词之间关系的词类,介词常与动词、形容词和名词搭配表示不同意义。介词短语中介词后接名词、代词或可以替代名词的词(如:动名词v-ing).介词后的代词永远为宾格形式。介词的种类: (1)简单介词:about, across, after, against, among, around, at, before, behind, below, beside, but, by, down, during, for, from, in, of, on, over, near, round, since, to, under, up, with等等。 (2)合成介词:inside, into, outside, throughout, upon, without, within (3)短语介词:according to, along with, apart from, because of, in front of, in spite of, instead of, owing to, up to, with reguard to (4)分词介词:considering, reguarding, including, concerning 介词短语:构成 介词+名词We go to school from Monday to Saturday. 介词+代词Could you look for it instead of me? 介词+动名词He insisted on staying home. 介词+连接代/副词I was thinking of how we could get there. 介词+不定式/从句He gives us some advice on how to finish it. 介词的用法: 一、介词to的常见用法 1.动词+to a)动词+ to adjust to适应, attend to处理;照料, agree to赞同,

超全的英语介词用法归纳总结

超全的英语介词用法归纳总结常用介词基本用法辨析 表示方位的介词:in, to, on 1. in 表示在某地范围之内。 Shanghai is/lies in the east of China. 上海在中国的东部。 2. to 表示在某地范围之外。 Japan is/lies to the east of China. 日本位于中国的东面。 3. on 表示与某地相邻或接壤。 Mongolia is/lies on the north of China. 蒙古国位于中国北边。 表示计量的介词:at, for, by 1. at 表示“以……速度”“以……价格”。 It flies at about 900 kilometers an hour. 它以每小时900公里的速度飞行。 I sold my car at a high price. 我以高价出售了我的汽车。 2. for 表示“用……交换,以……为代价”。 He sold his car for 500 dollars. 他以五百元把车卖了。 注意:at表示单价(price) ,for表示总钱数。

3. by 表示“以……计”,后跟度量单位。 They paid him by the month. 他们按月给他计酬。 Here eggs are sold by weight. 在这里鸡蛋是按重量卖的。 表示材料的介词:of, from, in 1. of 成品仍可看出原料。 This box is made of paper. 这个盒子是纸做的。 2. from 成品已看不出原料。 Wine is made from grapes. 葡萄酒是葡萄酿成的。 3. in 表示用某种材料或语言。 Please fill in the form in pencil first. 请先用铅笔填写这个表格。They talk in English. 他们用英语交谈。 表示工具或手段的介词:by, with, on 1. by 用某种方式,多用于交通。 I went there by bus. 我坐公共汽车去那儿。 2. with表示“用某种工具”。 He broke the window with a stone. 他用石头把玻璃砸坏了。注意:with表示用某种工具时,必须用冠词或物主代词。

to与for的用法和区别

to与for的用法和区别 一般情况下, to后面常接对象; for后面表示原因与目的为多。 Thank you for helping me. Thanks to all of you. to sb.表示对某人有直接影响比如,食物对某人好或者不好就用to; for表示从意义、价值等间接角度来说,例如对某人而言是重要的,就用for. for和to这两个介词,意义丰富,用法复杂。这里仅就它们主要用法进行比较。 1. 表示各种“目的” 1. What do you study English for? 你为什么要学英语? 2. She went to france for holiday. 她到法国度假去了。 3. These books are written for pupils. 这些书是为学生些的。 4. hope for the best, prepare for the worst. 作最好的打算,作最坏的准备。 2.对于 1.She has a liking for painting. 她爱好绘画。 2.She had a natural gift for teaching. 她对教学有天赋/ 3.表示赞成同情,用for不用to. 1. Are you for the idea or against it? 你是支持还是反对这个想法? 2. He expresses sympathy for the common people.. 他表现了对普通老百姓的同情。 3. I felt deeply sorry for my friend who was very ill. 4 for表示因为,由于(常有较活译法) 1 Thank you for coming. 谢谢你来。 2. France is famous for its wines. 法国因酒而出名。 5.当事人对某事的主观看法,对于(某人),对…来说(多和形容词连用)用介词to,不用for.. He said that money was not important to him. 他说钱对他并不重要。 To her it was rather unusual. 对她来说这是相当不寻常的。 They are cruel to animals. 他们对动物很残忍。 6.for和fit, good, bad, useful, suitable 等形容词连用,表示适宜,适合。 Some training will make them fit for the job. 经过一段训练,他们会胜任这项工作的。 Exercises are good for health. 锻炼有益于健康。 Smoking and drinking are bad for health. 抽烟喝酒对健康有害。 You are not suited for the kind of work you are doing. 7. for表示不定式逻辑上的主语,可以用在主语、表语、状语、定语中。 1.It would be best for you to write to him. 2.The simple thing is for him to resign at once. 3.There was nowhere else for me to go. 4.He opened a door and stood aside for her to pass.

(完整版)介词for用法归纳

介词for用法归纳 用法1:(表目的)为了。如: They went out for a walk. 他们出去散步了。 What did you do that for? 你干吗这样做? That’s what we’re here for. 这正是我们来的目的。 What’s she gone for this time? 她这次去干什么去了? He was waiting for the bus. 他在等公共汽车。 【用法说明】在通常情况下,英语不用for doing sth 来表示目的。如: 他去那儿看他叔叔。 误:He went there for seeing his uncle. 正:He went there to see his uncle. 但是,若一个动名词已名词化,则可与for 连用表目的。如: He went there for swimming. 他去那儿游泳。(swimming 已名词化) 注意:若不是表目的,而是表原因、用途等,则其后可接动名词。(见下面的有关用法) 用法2:(表利益)为,为了。如: What can I do for you? 你想要我什么? We study hard for our motherland. 我们为祖国努力学习。 Would you please carry this for me? 请你替我提这个东西好吗? Do more exercise for the good of your health. 为了健康你要多运动。 【用法说明】(1) 有些后接双宾语的动词(如buy, choose, cook, fetch, find, get, order, prepare, sing, spare 等),当双宾语易位时,通常用for 来引出间接宾语,表示间接宾语为受益者。如: She made her daughter a dress. / She made a dress for her daughter. 她为她女儿做了件连衣裙。 He cooked us some potatoes. / He cooked some potatoes for us. 他为我们煮了些土豆。 注意,类似下面这样的句子必须用for: He bought a new chair for the office. 他为办公室买了张新办公椅。 (2) 注意不要按汉语字面意思,在一些及物动词后误加介词for: 他们决定在电视上为他们的新产品打广告。 误:They decided to advertise for their new product on TV. 正:They decided to advertise their new product on TV. 注:advertise 可用作及物或不及物动词,但含义不同:advertise sth=为卖出某物而打广告;advertise for sth=为寻找某物而打广告。如:advertise for a job=登广告求职。由于受汉语“为”的影响,而此处误加了介词for。类似地,汉语中的“为人民服务”,说成英语是serve the people,而不是serve for the people,“为某人的死报仇”,说成英语是avenge sb’s death,而不是avenge for sb’s death,等等。用法3:(表用途)用于,用来。如: Knives are used for cutting things. 小刀是用来切东西的。 This knife is for cutting bread. 这把小刀是用于切面包的。 It’s a machine for slicing bread. 这是切面包的机器。 The doctor gave her some medicine for her cold. 医生给了她一些感冒药。 用法4:为得到,为拿到,为取得。如: He went home for his book. 他回家拿书。 He went to his friend for advice. 他去向朋友请教。 She often asked her parents for money. 她经常向父母要钱。

【备战高考】英语介词用法总结(完整)

【备战高考】英语介词用法总结(完整) 一、单项选择介词 1. passion, people won't have the motivation or the joy necessary for creative thinking. A.For . B.Without C.Beneath D.By 【答案】B 【解析】 【详解】 考查介词辨析。句意:没有激情,人们就不会有创新思维所必须的动机和快乐。A. For 对于;B. Without没有; C. Beneath在……下面 ; D. By通过。没有激情,人们就不会有创新思维所必须的动机和快乐。所以空处填介词without。故填without。 2.Modern zoos should shoulder more social responsibility _______ social progress and awareness of the public. A.in light of B.in favor of C.in honor of D.in praise of 【答案】A 【解析】 【分析】 【详解】 考查介词短语。句意:现代的动物园应该根据社会的进步和公众的意识来承担更多的社会责任。A. in light of根据,鉴于;B. in favor of有利于,支持;C. in honor of 为了纪念;D. in praise of歌颂,为赞扬。此处表示根据,故选A。 3.If we surround ourselves with people _____our major purpose, we can get their support and encouragement. A.in sympathy with B.in terms of C.in honour of D.in contrast with 【答案】A 【解析】 【详解】 考查介词短语辨析。句意:如果我们周围都是认同我们主要前进目标的人,我们就能得到他们的支持和鼓励。A. in sympathy with赞成;B. in terms of 依据;C. in honour of为纪念; D. in contrast with与…形成对比。由“we can get their support and encouragement”可知,in sym pathy with“赞成”符合句意。故选A项。 4.Elizabeth has already achieved success_____her wildest dreams. A.at B.beyond C.within D.upon

英语介词for的用法归纳总结.doc

英语介词for的用法归纳总结用法1:(介词for表目的)为了 They went out for a walk. 他们出去散步了。 What did you do that for? 你干吗这样做? That s what we re here for. 这正是我们来的目的。 What s she gone for this time? 她这次去干什么去了? He was waiting for the bus. 他在等公共汽车。 【用法说明】在通常情况下,英语不用for doing sth 来表示目的 他去那儿看他叔叔。 误:He went there for seeing his uncle. 正:He went there to see his uncle. 但是,若一个动名词已名词化,则可与for 连用表目的 He went there for swimming. 他去那儿游泳。(swimming 已名词化) 注意:若不是表目的,而是表原因、用途等,则其后可接动名词。(见下面的有关用法) 用法2:(介词for表利益)为,为了 What can I do for you? 你想要我什么? We study hard for our motherland. 我们为祖国努力学习。 Would you please carry this for me? 请你替我提这个东西好吗?

Do more exercise for the good of your health. 为了健康你要多运动。 【用法说明】(1) 有些后接双宾语的动词(如buy, choose, cook, fetch, find, get, order, prepare, sing, spare 等),当双宾语易位时,通常用for 来引出间接宾语,表示间接宾语为受益者 She made her daughter a dress. / She made a dress for her daughter. 她为她女儿做了件连衣裙。 He cooked us some potatoes. / He cooked some potatoes for us. 他为我们煮了些土豆。 注意,类似下面这样的句子必须用for: He bought a new chair for the office. 他为办公室买了张新办公椅。 (2) 注意不要按汉语字面意思,在一些及物动词后误加介词for: 他们决定在电视上为他们的新产品打广告。 误:They decided to advertise for their new product on TV. 正:They decided to advertise their new product on TV. 注:advertise 可用作及物或不及物动词,但含义不同:advertise sth=为卖出某物而打广告;advertise for sth=为寻找某物而打广告advertise for a job=登广告求职。由于受汉语为的影响,而此处误加了介词for。类似地,汉语中的为人民服务,说成英语是serve the people,而不是serve for the people,为某人的死报仇,说成英语是avenge sb s death,而不是avenge for sb s death,等等。 用法3:(介词for表用途)用于,用来 Knives are used for cutting things. 小刀是用来切东西的。

介词for用法完全归纳

用法1:(表目的)为了。如: They went out for a walk. 他们出去散步了。 What did you do that for? 你干吗这样做? That’s what we’re here for. 这正是我们来的目的。 What’s she gone for this time? 她这次去干什么去了? He was waiting for the bus. 他在等公共汽车。 【用法说明】在通常情况下,英语不用for doing sth 来表示目的。如:他去那儿看他叔叔。 误:He went there for seeing his uncle. 正:He went there to see his uncle. 但是,若一个动名词已名词化,则可与for 连用表目的。如: He went there for swimming. 他去那儿游泳。(swimming 已名词化) 注意:若不是表目的,而是表原因、用途等,则其后可接动名词。(见下面的有关用法) 用法2:(表利益)为,为了。如: What can I do for you? 你想要我什么? We study hard for our motherland. 我们为祖国努力学习。 Would you please carry this for me? 请你替我提这个东西好吗? Do more exercise for the good of your health. 为了健康你要多运动。 【用法说明】(1) 有些后接双宾语的动词(如buy, choose, cook, fetch, find, get, order, prepare, sing, spare 等),当双宾语易位时,通常用for 来引出间接宾语,表示间接宾语为受益者。如:

常用介词用法(for to with of)

For的用法 1. 表示“当作、作为”。如: I like some bread and milk for breakfast. 我喜欢把面包和牛奶作为早餐。 What will we have for supper? 我们晚餐吃什么? 2. 表示理由或原因,意为“因为、由于”。如: Thank you for helping me with my English. 谢谢你帮我学习英语。 3. 表示动作的对象或接受者,意为“给……”、“对…… (而言)”。如: Let me pick it up for you. 让我为你捡起来。 Watching TV too much is bad for your health. 看电视太多有害于你的健康。 4. 表示时间、距离,意为“计、达”。如: I usually do the running for an hour in the morning. 我早晨通常跑步一小时。 We will stay there for two days. 我们将在那里逗留两天。 5. 表示去向、目的,意为“向、往、取、买”等。如: Let’s go for a walk. 我们出去散步吧。 I came here for my schoolbag.我来这儿取书包。 I paid twenty yuan for the dictionary. 我花了20元买这本词典。 6. 表示所属关系或用途,意为“为、适于……的”。如: It’s time for school. 到上学的时间了。 Here is a letter for you. 这儿有你的一封信。 7. 表示“支持、赞成”。如: Are you for this plan or against it? 你是支持还是反对这个计划? 8. 用于一些固定搭配中。如: Who are you waiting for? 你在等谁? For example, Mr Green is a kind teacher. 比如,格林先生是一位心地善良的老师。 尽管for 的用法较多,但记住常用的几个就可以了。 to的用法: 一:表示相对,针对 be strange (common, new, familiar, peculiar) to This injection will make you immune to infection. 二:表示对比,比较 1:以-ior结尾的形容词,后接介词to表示比较,如:superior ,inferior,prior,senior,junior 2: 一些本身就含有比较或比拟意思的形容词,如equal,similar,equivalent,analogous A is similar to B in many ways.

for的用法完全归纳

for的用法完全归纳 用法1:(表目的)为了。如: They went out for a walk. 他们出去散步了。 What did you do that for? 你干吗这样做? That’s what we’re here for. 这正是我们来的目的。 What’s she gone for this time? 她这次去干什么去了? He was waiting for the bus. 他在等公共汽车。 在通常情况下,英语不用for doing sth 来表示目的。如:他去那儿看他叔叔。 误:He went there for seeing his uncle.正:He went there to see his uncle. 但是,若一个动名词已名词化,则可与for 连用表目的。如: He went there for swimming. 他去那儿游泳。(swimming 已名词化) 注意:若不是表目的,而是表原因、用途等,则其后可接动名词。 用法2:(表利益)为,为了。如: What can I do for you? 你想要我什么? We study hard for our motherland. 我们为祖国努力学习。 Would you please carry this for me? 请你替我提这个东西好吗? Do more exercise for the good of your health. 为了健康你要多运动。 (1)有些后接双宾语的动词(如buy, choose, cook, fetch, find, get, order, prepare, sing, spare 等),当双宾语易位时,通 常用for 来引出间接宾语,表示间接宾语为受益者。如: She made her daughter a dress. / She made a dress for her daughter. 她为她女儿做了件连衣裙。 He cooked us some potatoes. / He cooked some potatoes for us. 他为我们煮了些土豆。 注意,类似下面这样的句子必须用for: He bought a new chair for the office. 他为办公室买了张新办公椅。 (2) 注意不要按汉语字面意思,在一些及物动词后误加介词for: 他们决定在电视上为他们的新产品打广告。 误:They decided to advertise for their new product on TV. 正:They decided to advertise their new product on TV. 注:advertise 可用作及物或不及物动词,但含义不同:advertise sth=为卖出某物而打广告;advertise for sth=为寻找某物而打广告。如:advertise for a job=登广告求职。由于受汉语“为”的影响,而此处误加了介词for。类似地,汉语中的“为人民服务”,说成英语是serve the people,而不是serve for the people,“为某人的死报仇”,说成英语是avenge sb’s death,而不是avenge for sb’s death,等等。 用法3:(表用途)用于,用来。如: Knives are used for cutting things. 小刀是用来切东西的。 This knife is for cutting bread. 这把小刀是用于切面包的。 It’s a machine for slicing bread. 这是切面包的机器。 The doctor gave her some medicine for her cold. 医生给了她一些感冒药。 用法4:为得到,为拿到,为取得。如: He went home for his book. 他回家拿书。 He went to his friend for advice. 他去向朋友请教。 She often asked her parents for money. 她经常向父母要钱。 We all hope for success. 我们都盼望成功。 Are you coming in for some tea? 你要不要进来喝点茶? 用法5:给(某人),供(某人)用。如: That’s for you. 这是给你的。 Here is a letter for you. 这是你的信。 Have you room for me there? 你那边能给我腾出点地方吗? 用法6:(表原因、理由)因为,由于。如:

for和to区别

1.表示各种“目的”,用for (1)What do you study English for 你为什么要学英语? (2)went to france for holiday. 她到法国度假去了。 (3)These books are written for pupils. 这些书是为学生些的。 (4)hope for the best, prepare for the worst. 作最好的打算,作最坏的准备。 2.“对于”用for (1)She has a liking for painting. 她爱好绘画。 (2)She had a natural gift for teaching. 她对教学有天赋/ 3.表示“赞成、同情”,用for (1)Are you for the idea or against it 你是支持还是反对这个想法? (2)He expresses sympathy for the common people.. 他表现了对普通老百姓的同情。 (3)I felt deeply sorry for my friend who was very ill. 4. 表示“因为,由于”(常有较活译法),用for (1)Thank you for coming. 谢谢你来。

(2)France is famous for its wines. 法国因酒而出名。 5.当事人对某事的主观看法,“对于(某人),对…来说”,(多和形容词连用),用介词to,不用for. (1)He said that money was not important to him. 他说钱对他并不重要。 (2)To her it was rather unusual. 对她来说这是相当不寻常的。 (3)They are cruel to animals. 他们对动物很残忍。 6.和fit, good, bad, useful, suitable 等形容词连用,表示“适宜,适合”,用for。(1)Some training will make them fit for the job. 经过一段训练,他们会胜任这项工作的。 (2)Exercises are good for health. 锻炼有益于健康。 (3)Smoking and drinking are bad for health. 抽烟喝酒对健康有害。 (4)You are not suited for the kind of work you are doing. 7. 表示不定式逻辑上的主语,可以用在主语、表语、状语、定语中。 (1)It would be best for you to write to him. (2) The simple thing is for him to resign at once.

介词for 的常见用法归纳

介词for 的常见用法归纳 贵州省黔东南州黎平县黎平一中英语组廖钟雁介词for 用法灵活并且搭配能力很强,是一个使用频率非常高的词,也是 高考必考的重要词汇,现将其常见用法归纳如下,供参考。 1.表时间、距离或数量等。 ①意为“在特定时间,定于,安排在约定时间”。如: The meeting is arranged for 9 o’clock. 会议安排在九点进行。 ②意为“持续达”,常于last、stay 、wait等持续性动词连用,表动作持续的时间,有时可以省略。如: He stayed for a long time. 他逗留了很久。 The meeting lasted (for)three hours. 会议持续了三小时。 ③意为“(距离或数量)计、达”。例如: He walked for two miles. 他走了两英里。 The shop sent me a bill for $100.商店给我送来了100美元的账单。 2. 表方向。意为“向、朝、开往、前往”。常与head、leave 、set off、start 等动词连用。如: Tomorrow Tom will leave for Beijing. 明天汤姆要去北京。 He put on his coat and headed for the door他穿上大衣向门口走去。 介词to也可表示方向,但往往与come、drive 、fly、get、go、lead、march、move、return、ride、travel、walk等动词连用。 3.表示理由或原因,意为“因为、由于”。常与thank、famous、reason 、sake 等词连用。如: Thank you for helping me with my English. 谢谢你帮我学习英语。 For several reasons, I’d rather not meet him. 由于种种原因,我宁可不见他。 The West Lake is famous for its beautiful scenery.西湖因美景而闻名。 4.表示目的,意为“为了、取、买”等。如: Let’s go for a walk. 我们出去散步吧。 I came here for my schoolbag.我来这儿取书包。 He plays the piano for pleasure. 他弹钢琴是为了消遣。 There is no need for anyone to know. 没必要让任何人知道。 5.表示动作的对象或接受者,意为“给、为、对于”。如: Let me pick it up for you. 让我为你捡起来。 Watching TV too much is bad for your health. 看电视太多有害于你的健康。 Here is a letter for you. 这儿有你的一封信。

双宾语tofor的用法

1. 两者都可以引出间接宾语,但要根据不同的动词分别选用介词to 或for: (1) 在give, pass, hand, lend, send, tell, bring, show, pay, read, return, write, offer, teach, throw 等之后接介词to。 如: 请把那本字典递给我。 正:Please hand me that dictionary. 正:Please hand that dictionary to me. 她去年教我们的音乐。 正:She taught us music last year. 正:She taught music to us last year. (2) 在buy, make, get, order, cook, sing, fetch, play, find, paint, choose,prepare, spare 等之后用介词for 。如: 他为我们唱了首英语歌。 正:He sang us an English song. 正:He sang an English song for us. 请帮我把钥匙找到。 正:Please find me the keys. 正:Please find the keys for me. 能耽搁你几分钟吗(即你能为我抽出几分钟吗)? 正:Can you spare me a few minutes? 正:Can you spare a few minutes for me? 注:有的动词由于搭配和含义的不同,用介词to 或for 都是可能的。如: do sb a favou r do a favour for sb 帮某人的忙 do sb harnn= do harm to sb 对某人有害

英语介词的用法总结

介词的用法 1.表示地点位置的介词 1)at ,in, on, to,for at (1)表示在小地方; (2)表示“在……附近,旁边” in (1)表示在大地方; (2)表示“在…范围之内”。 on 表示毗邻,接壤,“在……上面”。 to 表示在……范围外,不强调是否接壤;或“到……” 2)above, over, on 在……上 above 指在……上方,不强调是否垂直,与below相对; over指垂直的上方,与under相对,但over与物体有一定的空间,不直接接触。 on表示某物体上面并与之接触。 The bird is flying above my head. There is a bridge over the river. He put his watch on the desk. 3)below, under 在……下面 under表示在…正下方 below表示在……下,不一定在正下方 There is a cat under the table. Please write your name below the line. 4)in front [frant]of, in the front of在……前面 in front of…意思是“在……前面”,指甲物在乙物之前,两者互不包括;其反义词是behind(在……的后面)。There are some flowers in front of the house.(房子前面有些花卉。) in the front of 意思是“在…..的前部”,即甲物在乙物的内部.反义词是at the back of…(在……范围内的后部)。 There is a blackboard in the front of our classroom. 我们的教室前边有一块黑板。 Our teacher stands in the front of the classroom. 我们的老师站在教室前.(老师在教室里) 5)beside,behind beside 表示在……旁边 behind 表示在……后面 2.表示时间的介词 1)in , on,at 在……时 in表示较长时间,如世纪、朝代、时代、年、季节、月及一般(非特指)的早、中、晚等。 如in the 20th century, in the 1950s, in 1989, in summer, in January, in the morning, in one’s life , in one’s thirties等。 on表示具体某一天及其早、中、晚。 如on May 1st, on Monday, on New Year’s Day, on a cold night in January, on a fine morning, on Sunday afternoon等。 at表示某一时刻或较短暂的时间,或泛指圣诞节,复活节等。 如at 3:20, at this time of year, at the beginning of, at the end of …, at the age of …, at Christmas,at night, at noon, at this moment等。 注意:在last, next, this, that, some, every 等词之前一律不用介词。如:We meet every day. 2)in, after 在……之后 “in +段时间”表示将来的一段时间以后; “after+段时间”表示过去的一段时间以后; “after+将来的时间点”表示将来的某一时刻以后。 3)from, since 自从…… from仅说明什么时候开始,不说明某动作或情况持续多久;

介词的归纳

介词的归纳 一、单项选择介词 1.(重庆)Last year was the warmest year on record, with global temperature 0.68 ℃ ________ the average. A.below B.on C.at D.above 【答案】D 【解析】 【详解】 考查介词。句意:去年是有纪录以来最热的一年,全球平均气温上升0.68度。A. below低于;B. on在……之上;C. at在;D. above超过,多于。根据前一句Last year was the warmest year on record推知,温度应该是上升了,故用介词above。 【点睛】 with的复合结构中,复合宾语中第一部分宾语由名词和代词充当,第二部分补足语由形容词,副词,介词短语,动词不定式或分词充当。而本题考查with +名词/代词+介词短语,而介词的使用则根据当时语境的提示来做出相应的变化即句中的the warmest year on record 起重要作用,可知高出平均气温。 2.According to Baidu, the high-quality content of Cloud Music will reach massive users _______ Baidu’s app and video platform. A.in honor of B.in view of C.by virtue of D.by way of 【答案】C 【解析】 【详解】 考查介词短语。句意:根据百度的说法,云音乐的高质量内容将借助于百度应用和视频平台到达广大用户。A. in honor of向……致敬;B. in view of考虑到;C. by virtue of借助于;D. by way of通过。根据句意可知,此处要表达“借助于”。故选C项。 3.We charge parcels ________ weight, rather than individual units. A.in honor of B.in contact with C.in terms of D.in connection with 【答案】C 【解析】 【详解】 考查介词短语。句意:我们根据包裹的重量,而不是包裹的件数收费。A. in honor of为了对……表示敬意;B. in contact with与……有联系,接触;C. in terms of根据,在……方面;D. in connection with与……有关,有联系。表示根据什么计费。故选C。 【点睛】

介词用法归纳

介词(preposition) 又称前置词,是一种虚词。介词不能单独做句子成分。介词后须接宾语,介词与其宾语构成介词短语。 一、介词从其构成来看可以分为: 1、简单介词(Simple prepositions)如:at ,by, for, in, from, since, through等; 2、复合介词(Compound prepositions)如:onto, out of, without, towards等; 3、短语介词(phrasal prepositions)如;because of, instead of, on account of, in spite of, in front of等; 4、二重介词(double prepositions)如:from behind, from under, till after等; 5、分词介词(participial prepositions),又可称动词介词(verbal prepositions)如:during, concerning, excepting, considering, past等。 二、常见介词的基本用法 1、 about 关于 Do you know something about Tom? What about this coat?(……怎么样) 2、 after 在……之后 I’m going to see you after supper. Tom looked after his sick mother yesterday.(照看) 3、 across 横过 Can you swim across the river. 4、 against 反对 Are you for or against me? Nothing could make me turn against my country.(背叛) 5、 along 沿着 We walked along the river bank. 6、 before 在……之前 I hope to get there before seven o’clock. It looks as though it will snow before long.(不久) 7、behind 在……后面 The sun is hidden behind the clouds. 8、by 到……时 We had learned ten English songs by the end of last term. 9、during 在……期间 Where are you going during the holiday. 10、except 除了 Everyone except you answered the question correctly. 11、for 为了 The students are studying hard for the people. 12、from 从 I come from Shanghai. 13、in 在……里 on 在……上面 under在……下面 There are two balls in/on/under the desk. 14、near 在……附近 We live near the park. 15、of ……的 Do you know the name of the winner. 16、over 在……正上方 There is a bridge over the river. Tom goes over his English every day.(复习) 17、round/around 围绕 The students stand around the teacher. 18、to 朝……方向 Can you tell me the way to the cinema. 19、towards朝着 The car is traveling towards Beijing.

相关文档