 Contents | Voicebox: Speech Processing Toolbox for MATLAB |
| activlev | ACTIVLEV Measure active speech level as in ITU-T P.56 [LEV,AF,FSO]=(sp,FS,MODE) |
| atan2sc | ATAN2SC sin and cosine of atan(y/x) [S,C,R,T]=(Y,X) |
| axisenlarge | AXISENLARGE - enlarge the axes of a figure (f,h) |
| bark2frq | BARK2FRQ Convert the BARK frequency scale to Hertz FRQ=(BARK) |
| bitsprec | BITSPREC round values to a specified fixed or floating precision (X,N,MODE) |
| cblabel | h is the handle of the colorbar, axis or figure |
| ccwarpf | CCWARPF Warp cepstral coefficients M=(F,N,S) |
| cep2pow | CEP2POW convert cepstral means and variances to the power domain |
| choosenk | CHOOSENK All choices of K elements taken from 1:N [X]=(N,K) |
| choosrnk | CHOOSRNK All choices of K elements taken from 1:N with replacement. [X]=(N,K) |
| correlogram | make correlogram, |
| distchar | DISTCHAR calculates the cosh spectral distance between AR coefficients D=(AR1,AR2,MODE) |
| distchpf | DISTCHPF calculates the cosh spectral distance between power spectra D=(PF1,PF2,MODE) |
| disteusq | DISTEUSQ calculate euclidean, squared euclidean or mahanalobis distance D=(X,Y,MODE,W) |
| distisar | DISTISAR calculates the Itakura-Saito distance between AR coefficients D=(AR1,AR2,MODE) |
| distispf | DISTISPF calculates the Itakura-Saito spectral distance between power spectra D=(PF1,PF2,MODE) |
| distitar | DISTITAR calculates the Itakura distance between AR coefficients D=(AR1,AR2,MODE) |
| distitpf | DISTITPF calculates the Itakura spectral distance between power spectra D=(PF1,PF2,MODE) |
| ditherq | DITHERQ add dither and quantize [Y,ZF]=(X,M,ZI) |
| dlyapsq | Solves the discrete Lyapunov equation AV'VA' - V'V +BB' =0 |
| dualdiag | DUALDIAG Simultaneous diagonalisation of two hermitian matrices [A,D,E]=(W,B) |
| dypsa | DYPSA Derive glottal closure instances from speech [gci,goi] = (s,fs) |
| enframe | ENFRAME split signal up into (overlapping) frames: one per row. [F,T]=(X,WIN,INC) |
| entropy | ENTROPY calculates the entropy of discrete and sampled continuous distributions H=(P,DIM,COND,ARG,STEP) |
| erb2frq | ERB2FRQ Convert ERB frequency scale to Hertz FRQ=(ERB) |
| estnoiseg | ESTNOISEG - estimate MMSE noise spectrum [x,zo]=(yf,tz,pp) |
| estnoisem | ESTNOISEM - estimate noise spectrum using minimum statistics |
| ewgrpdel | EWGRPDEL calculates the energy weighted group delay waveform Y=(X,W,M) |
| fig2emf | FIG2EMF save a figure in windows metafile format (H,S,P) |
| figbolden | embolden the current figure |
| filtbankm | FILTBANKM determine matrix for a linear/mel/erb/bark-spaced filterbank [X,MN,MX]=(P,N,FS,FL,FH,W) |
| filterbank | FILTERBANK appply filterbank to a signal: [y,zo]=(b,a,x,gd) |
| findpeaks | FINDPEAKS finds peaks with optional quadratic interpolation [K,V]=(X,M,W) |
| finishat | FINISHAT print estimated finish time of a long computation (FRAC,TOL,FMT) |
| fopenmkd | FOPENMKD is the same as FOPEN but creates any missing directories [fid,mes]=(fn,pe,mf,en) |
| frac2bin | FRAC2BIN Convert an column vector to binary S=(D,N,M) |
| fram2wav | FRAM2WAV converts frame values to a continuous waveform [W]=(X,TT,MODE) |
| frq2bark | FRQ2BARK Convert Hertz to BARK frequency scale BARK=(FRQ) |
| frq2erb | FRQ2ERB Convert Hertz to ERB frequency scale ERB=(FRQ) |
| frq2mel | FRQ2ERB Convert Hertz to Mel frequency scale MEL=(FRQ) |
| frq2midi | FRQ2MIDI Convert frequencies to musical note numbers [N,T]=(F) |
| fxpefac | FXPEFAC PEFAC pitch tracker [FX,TT,PV,FV]=(S,FS,TINC,M,PP) |
| fxrapt | FXRAPT RAPT pitch tracker [FX,VUV]=(S,FS) |
| gammabank | GAMMABANK gammatone filter bank [b,a,fx,bx,gd]=(n,fs,w,fc,bw,ph,k) |
| gausprod | GAUSPROD calculates a product of gaussians [G,U,K]=(M,C) |
| gaussmix | GAUSSMIX fits a gaussian mixture pdf to a set of data observations [m,v,w,g,f]=(x,c,l,m0,v0,w0,wx) |
| gaussmixd | GAUSSMIXD marginal and conditional Gaussian mixture densities |
| gaussmixg | GAUSSMIXG global mean, variance and mode of a GMM |
| gaussmixp | GAUSSMIXP calculate probability densities from a Gaussian mixture model |
| glotlf | GLOTLF Liljencrants-Fant glottal model U=(D,T,P) |
| glotros | GLOTROS Rosenberg glottal model U=(D,T,P) |
| gmmlpdf | GMMPDF calculated the pdf of a mixture of gaussians p=(x,m,v,w) |
| histndim | HISTNDIM - generates and/or plots an n-dimensional histogram |
| hostipinfo | HOSTIPINFO get host name and internet connection information |
| huffman | HUFFMAN calculates a D-ary huffman code [CC,LL]=(P,A) |
| imagehomog | IMAGEHOMOG Apply a homography transformation to an image with bilinear interpolation |
| importsii | IMPORTSII calculates the SII importance function per Hz or per Bark Q=(F,M) |
| irdct | IRDCT Inverse discrete cosine transform of real data X=(Y,N) |
| irfft | IRFFT Inverse fft of a conjugate symmetric spectrum X=(Y,N,D) |
| kmeanhar | KMEANS Vector quantisation using K-harmonic means algorithm [X,G,XN,GG]=(D,K,L,E,X0) |
| kmeanlbg | KMEANLBG Vector quantisation using the Linde-Buzo-Gray algorithm [X,ESQ,J]=(D,K) |
| kmeans | KMEANS Vector quantisation using K-means algorithm [X,ESQ,J]=(D,K,X0,L) |
| ldatrace | LDATRACE Calculates an LDA transform to maximize trace discriminant [a,f,B,W]=(b,w,n,c) |
| lin2pcma | LIN2PCMA Convert linear PCM to A-law P=(X,M,S) |
| lin2pcmu | LIN2PCMU Convert linear to Mu-law PCM P=(X,S) |
| lognmpdf | LOGNMPDF calculate pdf of a multivariate lognormal distribution P=(X,M,V) |
| logsum | LOGSUM logsum(x,d)=log(sum(exp(x).*k,d)) |
| lpcaa2ao | LPCAA2AO LPC: Convert area function to area ratios AO=(AA) |
| lpcaa2dl | LPCAA2DL LPC: Convert area coefficients to dct of log area DL=(AA) |
| lpcaa2rf | LPCAA2RF LPC: Convert vocal tract areas to reflection coefficients RF=(AA) |
| lpcao2rf | LPCAO2RF LPC: Convert area ratios to reflection coefficients RF=(AO) |
| lpcar2am | LPCAR2AM Convert ar coefs to ar coef matrix [AM,EM]=(AR,P) |
| lpcar2cc | LPCAR2CC LPC: Convert ar filter to complex cepstrum CC=(AR,NP) |
| lpcar2db | LPCAR2DB LPC: Convert AR coefs to power spectrum in dB DB=(AR) |
| lpcar2ff | LPCAR2FF LPC: Convert AR coefs to complex spectrum FF=(AR,NP) |
| lpcar2fm | LPCAR2RF Convert autoregressive coefficients to formant freq+amp+bw [N,F,A,B]=(AR,T) |
| lpcar2im | LPCAR2IM Convert AR coefs to impulse response IM=(AR,NP) |
| lpcar2ls | LPCAR2LS convert ar polynomial to line spectrum pair frequencies LS=(AR) |
| lpcar2pf | LPCAR2PF Convert AR coefs to power spectrum PF=(AR,NP) |
| lpcar2pp | LPCAR2PP LPC: Convert ar filter to power spectrum polynomial in cos(w) PP=(AR) |
| lpcar2ra | LPCAR2RA Convert ar filter to inverse filter autocorrelation coefs. RA=(AR) |
| lpcar2rf | LPCAR2RF Convert autoregressive coefficients to reflection coefficients AR=(RF) |
| lpcar2rr | LPCAR2RR Convert autoregressive coefficients to autocorrelation coefficients RR=(AR,P) |
| lpcar2zz | LPCAR2ZZ Convert ar filter to z-plane poles ZZ=(AR) |
| lpcauto | LPCAUTO performs autocorrelation LPC analysis [AR,E,K]=(S,P,T) |
| lpcbwexp | LPCBWEXP expand formant bandwidths of LPC filter ARX=(AR,BW) |
| lpccc2ar | LPCCC2AR Convert complex cepstrum to ar coefficients AR=(CC) |
| lpccc2cc | LPCCC2PF Extrapolate complex cepstrum C=(CC) |
| lpccc2db | LPCCC2DBf Convert complex cepstrum to dB power spectrum DB=(CC,NP,NC) |
| lpccc2pf | LPCCC2PF Convert complex cepstrum to power spectrum PF=(CC,NP,NC) |
| lpcconv | LPCCONV(from,to,x,y)->s convert between LPC parameter sets |
| lpccovar | LPCCOVAR performs covariance LPC analysis [AR,E,DC]=(S,P,T) |
| lpccw2zz | LPCPZ2ZZ LPC: Power spectrum roots to LPC poles ZZ=(CW) |
| lpcdb2pf | LPCDB2PF Convert decibel power spectrum to power spectrum PF=(DB) |
| lpcdl2aa | LPCDL2AA dct of log area to area coefficients AA=(DL) |
| lpcff2pf | LPCFF2PF Convert complex spectrum to power spectrum PF=(FF) |
| lpcfq2zz | LPCFQ2ZZ Convert frequencies and q factors to z-plane poles ZZ=(F,Q) |
| lpcifilt | LPCIFILT Apply inverse filter to speech signal U=(S,AR,T,DC,FADE) |
| lpcim2ar | LPCIM2AR Convert impulse response to AR coefs AR=(IM) |
| lpcis2rf | LPCRF2IS Convert inverse sines to reflection coefficients RF=(IS) |
| lpcla2rf | LPCLA2RF Convert log areas to reflection coefficients RF=(LA) |
| lpclo2rf | LPCLO2RF Convert log area ratios to reflection coefficients RF=(LO) |
| lpcls2ar | LPCLS2AR convert line spectrum pair frequencies to ar polynomial AR=(LS) |
| lpcpf2cc | LPCPF2CC Convert power spectrum to complex cepstrum CC=(PF,NP) |
| lpcpf2rr | LPCPF2RR convert power spectrum to autocorrelation coefs RR=(PF,P) |
| lpcpp2cw | LPCPP2PZ LPC: Convert power spectrum polynomial in cos(w) to power spectrum zeros CW=(RP) |
| lpcpp2pz | LPCPP2PZ LPC: Convert power spectrum polynomial in cos(w) to power spectrum zeros PZ=(RP) |
| lpcpz2zz | LPCPZ2ZZ LPC: Power spectrum roots to LPC poles ZZ=(PZ) |
| lpcra2pf | LPCAR2PF Convert AR coefs to power spectrum PF=(RA,NP) |
| lpcra2pp | LPCAR2PP LPC: Convert ar filter autocorrelation to power spectrum polynomial in cos(w) PP=(RA) |
| lpcrand | generate n random stable polynomials of order p with a minimum pole |
| lpcrf2aa | LPCRF2AA Convert reflection coefficients to area function AA=(RF) |
| lpcrf2ao | LPCRF2AO Convert reflection coefficients to area ratios AO=(RF) |
| lpcrf2ar | LPCRF2AR Convert reflection coefs to autoregressive coefs [AR,ARP,ARU,G]=(RF) |
| lpcrf2is | LPCRF2IS Convert reflection coefficients to inverse sines IS=(RF) |
| lpcrf2la | LPCRF2LA Convert reflection coefficients to log areas LA=(RF) |
| lpcrf2lo | LPCRF2LO Convert reflection coefficients to log area ratios LO=(RF) |
| lpcrf2rr | LPCRR2AR convert reflection coefs to autocorrelation coefs [RR,AR]=(RF,P) |
| lpcrr2am | LPCRR2AM Convert autocorrelation coefs to ar coef matrix [AM,EM]=(RR) |
| lpcrr2ar | LPCRR2AR convert autocorrelation coefs to ar coefs [AR,E]=(RR) |
| lpcss2zz | LPCSS2ZZ Convert s-place poles to z-plane poles ZZ=(SS) |
| lpczz2ar | LPCZZ2AR Convert z-place poles to ar coefficients AR=(ZZ) |
| lpczz2cc | LPCZZ2CC Convert poles to "complex" cepstrum CC=(ZZ,NP) |
| lpczz2ss | LPCZZ2SS Convert z-place poles to s-plane poles SS=(ZZ) |
| m2htmlpwd | M2HTMLPWD - create html documentation of files in current directory |
| maxfilt | MAXFILT find max of an exponentially weighted sliding window [Y,K,Y0]=(X,F,N,D,X0) |
| maxgauss | MAXGAUSS determine gaussian approximation to max of a gaussian vector [p,u,v,r]=(m,c,d) |
| meansqtf | AVEPSPEC calculates the mean square transfer function for a filter D=(B,A) |
| mel2frq | MEL2FRQ Convert Mel frequency scale to Hertz FRQ=(MEL) |
| melbankm | MELBANKM determine matrix for a mel/erb/bark-spaced filterbank [X,MN,MX]=(P,N,FS,FL,FH,W) |
| melcepst | MELCEPST Calculate the mel cepstrum of a signal C=(S,FS,W,NC,P,N,INC,FL,FH) |
| midi2frq | MIDI2FRQ Convert musical note numbers to frequencies F=(N,S) |
| minspane | MINSPANE calculate minimum spanning tree using euclidean distance [p,s]=X |
| mintrace | MINTRACE find row permutation to minimize the trace p=(x) |
| modspect | MODSPECT Calculate the modulation spectrum of a signal C=(S,FS,W,NC,P,N,INC,FL,FH) |
| momfilt | MOMFILT calculates moments of a signal using a sliding window Y=(X,R,W,M) |
| mos2pesq | MOS2PESQ convert MOS speech quality scores to PESQ p=(m) |
| nearnonz | NEARNONZ replace each zero element with the nearest non-zero element [V,Y,W]=nearnonz(X,D) |
| overlapadd | OVERLAPADD join overlapping frames together X=(F,WIN,INC) |
| pcma2lin | PCMU2LIN Convert A-law PCM to linear X=(P,M,S) |
| pcmu2lin | PCMU2LIN Convert Mu-law PCM to linear X=(P,S) |
| peak2dquad | PEAK2DQUAD find quadratically-interpolated peak in a 2D array |
| permutes | PERMUTES All N! permutations of 1:N + signatures [P,S]=(N) |
| pesq2mos | PESQ2MOS convert PESQ speech quality scores to MOS m=(p) |
| polygonarea | POLYGONAREA Calculate the area of a polygon |
| polygonwind | POLYGONWIND Test if points are inside a polygon |
| polygonxline | POLYGONXLINE Find where a line crosses a polygon [xc,ec,tc,xy0]=(p,l) |
| potsband | POTSBAND Design filter for 300-3400 telephone bandwidth [B,A]=(FS) |
| pow2cep | CEP2POW convert cepstral means and variances to the power domain |
| psycdigit | PSYCDIGIT measures psychometric function using TIDIGITS stimuli |
| psycest | Estimate multiple psychometric functions |
| psycestu | psycestu estimate unimodal psychometric function |
| psychofunc | Calculate psychometric functions: trial success probability versus SNR |
| qrabs | QRABS absolute value and normalization of a real quaternions [m,q]=[q1] |
| qrdivide | QRDIVIDE divdes two real quaternions q=[q1,q2] |
| qrdotdiv | QRDOTDIV divides two real quaternions arrays elementwise q=[x,y] |
| qrdotmult | QRDOTMULT multiplies together two real quaternions arrays q=[q1,q2] |
| qrmult | QRMULT multiplies together two real quaternions matrices q=[q1,q2] |
| qrpermute | QRPERMUTE transpose or permute a quaternion array y=[x,p] |
| quadpeak | PEAK2DQUAD find quadratically-interpolated peak in a N-D array |
| randfilt | RANDFILT Generate filtered gaussian noise without initial transient |
| randiscr | RANDISCR Generate discrete random numbers with specified probabiities [X]=(P,N,A) |
| randvec | RANDVEC Generate real or complex random vectors X=(N,M,C,W,MODE) |
| rdct | RDCT Discrete cosine transform of real data Y=(X,N,A,B) |
| readaif | READAIF Read a .AIF format sound file [Y,FS,WMODE,FIDX]=(FILENAME,MODE,NMAX,NSKIP) |
| readau | READAU Read a SUN .AU format sound file [Y,FS,H]=(FILENAME) |
| readcnx | READCNX Read a .CNX format sound file [Y,FS,H]=(FILENAME) |
| readflac | READWAV Read a .FLAC format sound file [Y,FS]=(FILENAME,MODE) |
| readhtk | READHTK read an HTK parameter file [D,FP,DT,TC,T]=(FILE) |
| readsfs | READSFS Read a .SFS format sound file [Y,FS,HD,FFX]=(FF,TY,SUB,MODE,NMAX,NSKIP,XPATH) |
| readsph | READSPH Read a SPHERE/TIMIT format sound file [Y,FS,WRD,PHN,FFX]=(FILENAME,MODE,NMAX,NSKIP) |
| readwav | READWAV Read a .WAV format sound file [Y,FS,WMODE,FIDX]=(FILENAME,MODE,NMAX,NSKIP) |
| rectifyhomog | RECTIFYHOMOG Apply rectifying homographies to an image set |
| rfft | RFFT Calculate the DFT of real data Y=(X,N,D) |
| rhartley | RHARTLEY Calculate the Hartley transform of real data Y=(X,N) |
| rnsubset | RNSUBSET choose k distinct random integers from 1:n M=(K,N) |
| rotation | ROTATION Encode and decode rotation matrices |
| rotax2qr | ROTQR2AX converts a rotation axis and angle to the corresponding real quaternion |
| roteu2qr | ROTEU2QR converts a sequence of Euler angles to a real unit quaternion |
| roteu2ro | ROTEU2QR converts a sequence of Euler angles to a real unit quaternion |
| rotmc2qc | ROTMC2QC converts a matrix of complex quaternion matrices to a matrix of complex quaternion vectors |
| rotmr2qr | ROTMR2QR converts a matrix of real quaternion matrices to quaternion vectors |
| rotpl2ro | ROTPL2RO find matrix to rotate in the plane containing u and v r=[u,v,t] |
| rotqc2mc | ROTQC2MC converts a matrix of complex quaternion vectors to quaternion matrices |
| rotqc2qr | ROTQC2QR converts a matrix of complex quaternion row vectors into real form |
| rotqr2ax | ROTQR2AX converts a real quaternion to the corresponding rotation axis and angle |
| rotqr2eu | ROTQR2EQ converts a real unit quaternion into the corresponding euler angles |
| rotqr2mr | ROTQR2MR converts a matrix of real quaternion vectors to quaternion matrices |
| rotqr2qc | ROTQR2QC converts a matrix of real quaternion vectors into complex form |
| rotqr2ro | ROTQR2RO converts a real quaternion to a 3x3 rotation matrix |
| rotqrmean | ROTQRMEAN calculates the mean rotation of a quaternion array [y,s]=[q] |
| rotqrvec | ROTQRVEC applies a quaternion rotation ot a vector array y=[q,x] |
| rotro2eu | ROTRO2EQ converts a 3x3 rotation matrix into the corresponding euler angles |
| rotro2pl | ROTRO2PL find the plane and rotation angle of a rotation matrix [u,v,t]=r |
| rotro2qr | ROTRO2QR converts a 3x3 rotation matrix to a real quaternion |
| rsfft | RSFFT fft of a real symmetric spectrum X=(Y,N) |
| sapisynth | SAPISYNTH text-to-speech synthesize of text string or matrix [X,FS,TXT]=(T,M) |
| schmitt | Pass input signal X through a schmitt trigger |
| sigalign | SIGALIGN align a clean reference with a noisy signal [d,g,rr,ss]=(s,r,maxd,m,fs) |
| sigma | Singularity in EGG by Multiscale Analysis (SIGMA) Algorithm |
| skew3d | SKEW3D Convert between a vector and the corresponding skew-symmetric matrix |
| snrseg | SNRSEG Measure segmental and global SNR [SEG,GLO]=(S,R,FS,M,TF) |
| soundspeed | SOUNDSPEED gives the speed of sound, density of air and acostuc impedance as a function of temp & pressure [V,D,Z]=(T,P,M,G) |
| specsub | SPECSUB performs speech enhancement using spectral subtraction [SS,ZO]=(S,FSZ,P) |
| specsubm | SPECSUBM obsolete speech enhancement algorithm - use specsub instead |
| spgrambw | SPGRAMBW Draw spectrogram [T,F,B]=(s,fs,mode,bw,fmax,db,tinc,ann) |
| sphrharm | SPHRHARM forward and inverse spherical harmonic transform |
| sprintsi | SPRINTSI Print X with SI multiplier S=(X,D,W) |
| ssubmmse | SSUBMMSE performs speech enhancement using mmse estimate of spectral amplitude or log amplitude [SS,ZO]=(S,FSZ,P) |
| stdspectrum | STDSPECTRUM Generate standard acoustic/speech spectra in s- or z-domain [B,A]=(S,M,F,N,ZI,BS,AS) |
| teager | TEAGER calculate teager energy waveform Y=(X,D,M) |
| txalign | TXALIGN Find the best alignment of two sets of time markers [KX,KY,N,M,S]=(X,Y,MAXT) |
| unixwhich | UNIXWHICH Search system path for an executable program [F]=(C,E) |
| upolyhedron | UPOLYHEDRON calculate uniform polyhedron characteristics |
| usasi | USASI generates N samples of USASI noise at sample frequency FS X=(N,FS) |
| v_colormap | V_COLORMAP set and plot color map |
| vadsohn | VADSOHN implements a voice activity detector [VS,ZO]=(S,FSZ,M,P) |
| voicebox | VOICEBOX set global parameters for Voicebox functions Y=(FIELD,VAL) |
| windinfo | WINDINFO window information and figures of merit X=(W,FS) |
| windows | WINDOWS Generate a standard windowing function (TYPE,N,MODE,P) |
| winenvar | WINENVAR get windows environment variable [D]=(N) |
| writehtk | WRITEHTK write data in HTK format []=(FILE,D,FP,TC) |
| writewav | WRITEWAV Creates .WAV format sound files FIDX=(D,FS,FILENAME,MODE,NSKIP,MASK) |
| xticksi | XTIXKSI labels the x-axis of a plot using SI multipliers S=(AH) |
| xyzticksi | XYZTIXKSI labels an axis of a plot using SI multipliers S=(AX,AH) |
| yticksi | YTIXKSI labels the y-axis of a plot using SI multipliers S=(AH) |
| zerocros | ZEROCROS finds the zeros crossings in a signal [T,S]=(X,M)% find zero crossings in a signal |
| zerotrim | ZEROTRIM Remove zero trailing rows and columns Z=(X) |
| zoomfft | ZOOMFFT DFT evaluated over a linear frequency range Y=(X,N,M,S,D) |
Master index