Formant Estimation and Tracking Using Deep Learning

Dissen, Yehoshua; Keshet, Joseph

doi:10.21437/interspeech.2016-490

Cited by 12 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…"KARMA" denotes the state-of-the-art KF-based formant tracking method published in [14]. "DeepF" (DeepFormants) denotes the deep-learning based formant tracking method proposed recently in [16], [18], [51]. It is worth emphasizing that DeepF is based on supervised learning and calls for an annotated speech corpus to be trained.…”

Section: E Experiments On Natural Speech Datamentioning

confidence: 99%

“…However, it should be mentioned here that there are a few exceptions, such as [15], which uses a non-negative matrix factorization (NMF)based source-filter modeling of speech signals. Recently, deep learning-based techniques [16]- [18] have also been studied as alternatives to conventional statistical signal processing-based formant estimation and tracking methods. These methods, however, are based on supervised machine learning, which calls for having annotated speech corpora with which to obtain the ground truth formant frequencies for system training.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Time-varying quasi-closed-phase analysis for accurate formant tracking in speech signals

Gowda

Kadiri

Story

et al. 2020

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

In this paper, we propose a new method for the accurate estimation and tracking of formants in speech signals using time-varying quasi-closed-phase (TVQCP) analysis. Conventional formant tracking methods typically adopt a two-stage estimate-and-track strategy wherein an initial set of formant candidates are estimated using short-time analysis (e.g., 10-50 ms), followed by a tracking stage based on dynamic programming or a linear state-space model. One of the main disadvantages of these approaches is that the tracking stage, however good it may be, cannot improve upon the formant estimation accuracy of the first stage. The proposed TVQCP method provides a singlestage formant tracking that combines the estimation and tracking stages into one. TVQCP analysis combines three approaches to improve formant estimation and tracking: (1) it uses temporally weighted quasi-closed-phase analysis to derive closed-phase estimates of the vocal tract with reduced interference from the excitation source, (2) it increases the residual sparsity by using the L1 optimization and (3) it uses time-varying linear prediction analysis over long time windows (e.g., 100-200 ms) to impose a continuity constraint on the vocal tract model and hence on the formant trajectories. Formant tracking experiments with a wide variety of synthetic and natural speech signals show that the proposed TVQCP method performs better than conventional and popular formant tracking tools, such as Wavesurfer and Praat (based on dynamic programming), the KARMA algorithm (based on Kalman filtering), and DeepFormants (based on deep neural networks trained in a supervised manner). Matlab scripts for the proposed method can be found at: https://github. com/njaygowda/ftrack

show abstract

Section: E Experiments On Natural Speech Datamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Time-varying quasi-closed-phase analysis for accurate formant tracking in speech signals

Gowda

Kadiri

Story

et al. 2020

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

show abstract

“…The aforementioned ad-hoc signal processing methods [17] usually emerge false peaks and formant merging when affected by high pitch or coarticulation. These problems can be alleviated by visually correcting with the help of linguistic knowledge and spectral analysis.…”

Section: Introductionmentioning

confidence: 99%

“…For example, Mehta et al evaluated their proposed Kalman-based autoregressive moving average modeling methods on this database [9]. Inspired by the great success of deep learning in many application areas, Dissen et al employed Long Short-Term Memory (LSTM) networks to train a supervised regression model between LPCCs plus Pitch-Synchronous Cepstrum Coefficients (named PSCCs) and handcorrected formant frequencies for every speech frame [17]. Later, Dissen et al [19] explored the potential of raw spectrograms (55 × 50 PSCCs) for formant tracking with Convolutional LSTM networks [20] and found that incorporating the PSCCs and LPCCs achieved the better general performance than using them separately.…”

Section: Introductionmentioning

confidence: 99%

Formant Tracking Using Dilated Convolutional Networks Through Dense Connection with Gating Mechanism

et al. 2020

View full text Add to dashboard Cite

Formant tracking is one of the most fundamental problems in speech processing. Traditionally, formants are estimated using signal processing methods. Recent studies showed that generic convolutional architectures can outperform recurrent networks on temporal tasks such as speech synthesis and machine translation. In this paper, we explored the use of Temporal Convolutional Network (TCN) for formant tracking. In addition to the conventional implementation, we modified the architecture from three aspects. First, we turned off the "causal" mode of dilated convolution, making the dilated convolution see the future speech frames. Second, each hidden layer reused the output information from all the previous layers through dense connection. Third, we also adopted a gating mechanism to alleviate the problem of gradient disappearance by selectively forgetting unimportant information. The model was validated on the open access formant database VTR. The experiment showed that our proposed model was easy to converge and achieved an overall mean absolute percent error (MAPE) of 8.2% on speech-labeled frames, compared to three competitive baselines of 9.4% (LSTM), 9.1% (Bi-LSTM) and 8.9% (TCN).

show abstract

“…ity [22][23][24][25][26]. Our methods rely on several advances over existing computational systems: novel representations of the speech signal and new structured prediction and deep learning algorithms.…”

Section: Introductionmentioning

confidence: 99%

Automatic Measurement of Pre-Aspiration

et al. 2017

Self Cite

View full text Add to dashboard Cite

Pre-aspiration is defined as the period of glottal friction occurring in sequences of vocalic/consonantal sonorants and phonetically voiceless obstruents. We propose two machine learning methods for automatic measurement of pre-aspiration duration: a feedforward neural network, which works at the frame level; and a structured prediction model, which relies on manually designed feature functions, and works at the segment level. The input for both algorithms is a speech signal of an arbitrary length containing a single obstruent, and the output is a pair of times which constitutes the pre-aspiration boundaries. We train both models on a set of manually annotated examples. Results suggest that the structured model is superior to the frame-based model as it yields higher accuracy in predicting the boundaries and generalizes to new speakers and new languages. Finally, we demonstrate the applicability of our structured prediction algorithm by replicating linguistic analysis of pre-aspiration in Aberystwyth English with high correlation.

show abstract

Formant Estimation and Tracking Using Deep Learning

Cited by 12 publications

References 25 publications

Time-varying quasi-closed-phase analysis for accurate formant tracking in speech signals

Time-varying quasi-closed-phase analysis for accurate formant tracking in speech signals

Formant Tracking Using Dilated Convolutional Networks Through Dense Connection with Gating Mechanism

Automatic Measurement of Pre-Aspiration

Contact Info

Product

Resources

About