COVAREP

Contributions

(this list corresponds to the latest release)

• Amplitude
• Discrete All-Pole (DAP) (env_dap)
• "True-Envelope" (TE) (env_te)
• Stabilized weighted linear prediction using short-time-energy (SWLP) (env_swlp_ste)
• Short-time-energy-weighted linear prediction (WLP) (env_wlp_ste)
• Extended weighted linear prediction using absolute-value-sum weighting (AVS-XLP) (env_xlp_avs)
• Compression using Frequency Warped CEPpstrum (FWCEP) (similar to MCEP) (hspec2fwcep, fwcep2hspec)
• Fast Inter-Harmonic Reconstruction (FIHR) and using the zero-phase speech signal (FIHRZP) (env_fihr, env_fihrzp)
• Discrete Cepstral Envelope (DCE) (env_dce_sfa)
• Discrete Cepstral Envelope - Multi Frame Analysis (DCE-MFA) (env_dce_mfa)
• Derivated Linear Interpolated Envelope (dLI-MFA) (env_dli_mfa)
• Phase
• Relative Phase Shift (RPS) (phase_rpspd)
• Phase Distortion (PD) (phase_rpspd)
• Phase Distortion Mean and Deviation (PDM and PDD) (hmpd_analysis)

• Chirp Group Delay formant tracking (CGD) (formant_CGDZP)
• Vowel space ratio of a speaker with respect to his/her reference population. (getVowelSpace)

• Peak Picking (sin_analysis)
• Harmonic Model (HM) (sin_analysis)
• Quasi-Harmonic Model (QHM) (partly) (sin_analysis)
• Adaptive Harmonic Model and Adaptive Iterative Refinement (aHM) (sin_analysis, ahm_air_analysis2)
• Harmonic synthesis (hm_synthesis4)
• Sinusoidal synthesis using OverLap-Add (OLA) (sin_synthesis_ola)

• Fan-Chirp Transform (FChT) (fcht, fchtgram)
• Phase unwrapping (T.Drugman) (drugmanunwrapping)

• Voice Activity Detection (VAD) (VAD)

• Polarity detection (RESKEW) (polarity_reskew)
• Pitch tracker using Summation of the Residual Harmonics (SRH) (pitch_srh)
• Speech Event Detection based on the Residual Excitation And Mean-based Signal (SEDREAMS) for Glottal Closure Instant (GCI) determination (gci_sedreams)
• SEDREAMS-based GCI method optimised for non-modal voice qualities (SE-VQ) (se_vq, se_vq_varf0)
• Detection of creaky voice (vocal/glottal fry) (detect_creaky_voice)
• Maximum voiced frequency estimation (maximum_voiced_frequency)
• Parameters
• Spectral tilt correlate (PEAKSLOPE) (peakslope)
• Maxima Dispersion Quotient (MDQ) (mdq)
• Normalised amplitude quotient (NAQ) (get_vq_params)
• Quasi-open quotient (QOQ) (get_vq_params)
• Difference in amplitude of the first two harmonics of the differentiated glottal source spectrum (H1-H2) (get_vq_params)
• Harmonic Richness Factor (HRF) (get_vq_params)
• Parabolic Spectral Parameter (PSP) (get_vq_params)
• Cepstral Peak Prominence (CPP) (cpp)
• Glottal flow estimation
• Iterative Adaptive Inverse Filtering (IAIF) Glottal Inverse Filtering (iaif, iaif_gci, iaif_ola)
• Glottal flow derivative estimation based on the complex cesptrum (complex_cepstrum)
• Glottal model based analysis
• Spectrum of the Liljencrants-Fant (LF) glottal flow derivative model (gfm_spec_lf)
• Rd parameter estimation of the LF glottal model based on Mean Squared Phase and 2nd order Difference (MSPD2) (rd_msp)
• Electroglottographic (EGG) analysis
• Open quotient measurement with number of peaks at glottal opening and closing (DECOM) (decom)
• Open quotient measurement (oq_egg)
• Open quotient measurement and Closure Peak Amplitude (PEAKDET) (peakdet)

• Harmonic Model + Phase Distortion (HMPD) vocoder (HMPD)

• Features for Robust DNN-based ASR (VAD_Drugman_FeatureExtraction)
• Modified Group Delay (modified_group_delay_feature)
• Automatic extraction of features available in COVAREP (COVAREP_feature_extraction)