Simulating Articulatory Trajectories with Phonological Feature Interpolation

As a first step towards a complete computational model of speech learning involving perception-production loops, we investigate the forward mapping between pseudo-motor commands and articulatory trajectories. Two phonological feature sets, based respectively on generative and articulatory phonology, are used to encode a phonetic target sequence. Different interpolation techniques are compared to generate smooth trajectories in these feature spaces, with a potential optimisation of the target value and timing to capture co-articulation effects. We report the Pearson correlation between a linear projection of the generated trajectories and articulatory data derived from a multi-speaker dataset of electromagnetic articulography (EMA) recordings. A correlation of 0.67 is obtained with an extended feature set based on generative phonology and a linear interpolation technique. We discuss the implications of our results for our understanding of the dynamics of biological motion.

Fill in the Gap! Combining Self-supervised Representation Learning with Neural Audio Synthesis for Speech Inpainting

Most speech self-supervised learning (SSL) models are trained with a pretext task which consists in predicting missing parts of the input signal, either future segments (causal prediction) or segments masked anywhere within the input (non-causal prediction). Learned speech representations can then be efficiently transferred to downstream tasks (e.g., automatic speech or speaker recognition). In the present study, we investigate the use of a speech SSL model for speech inpainting, that is reconstructing a missing portion of a speech signal from its surrounding context, i.e., fulfilling a downstream task that is very similar to the pretext task. To that purpose, we combine an SSL encoder, namely HuBERT, with a neural vocoder, namely HiFiGAN, playing the role of a decoder. In particular, we propose two solutions to match the HuBERT output with the HiFiGAN input, by freezing one and fine-tuning the other, and vice versa. Performance of both approaches was assessed in single- and multi-speaker settings, for both informed and blind inpainting configurations (i.e., the position of the mask is known or unknown, respectively), with different objective metrics and a perceptual evaluation. Performances show that if both solutions allow to correctly reconstruct signal portions up to the size of 200ms (and even 400ms in some cases), fine-tuning the SSL encoder provides a more accurate signal reconstruction in the single-speaker setting case, while freezing it (and training the neural vocoder instead) is a better strategy when dealing with multi-speaker data.

Investigating the dynamics of hand and lips in French Cued Speech using attention mechanisms and CTC-based decoding

Hard of hearing or profoundly deaf people make use of cued speech (CS) as a communication tool to understand spoken language. By delivering cues that are relevant to the phonetic information, CS offers a way to enhance lipreading. In literature, there have been several studies on the dynamics between the hand and the lips in the context of human production. This article proposes a way to investigate how a neural network learns this relation for a single speaker while performing a recognition task using attention mechanisms. Further, an analysis of the learnt dynamics is utilized to establish the relationship between the two modalities and extract automatic segments. For the purpose of this study, a new dataset has been recorded for French CS. Along with the release of this dataset, a benchmark will be reported for word-level recognition, a novelty in the automatic recognition of French CS.

BERT, can HE predict contrastive focus? Predicting and controlling prominence in neural TTS using a language model

Several recent studies have tested the use of transformer language model representations to infer prosodic features for text-to-speech synthesis (TTS). While these studies have explored prosody in general, in this work, we look specifically at the prediction of contrastive focus on personal pronouns. This is a particularly challenging task as it often requires semantic, discursive and/or pragmatic knowledge to predict correctly. We collect a corpus of utterances containing contrastive focus and we evaluate the accuracy of a BERT model, finetuned to predict quantized acoustic prominence features, on these samples. We also investigate how past utterances can provide relevant information for this prediction. Furthermore, we evaluate the controllability of pronoun prominence in a TTS model conditioned on acoustic prominence features.

Self-supervised speech unit discovery from articulatory and acoustic features using VQ-VAE

The human perception system is often assumed to recruit motor knowledge when processing auditory speech inputs. Using articulatory modeling and deep learning, this study examines how this articulatory information can be used for discovering speech units in a self-supervised setting. We used vector-quantized variational autoencoders (VQ-VAE) to learn discrete representations from articulatory and acoustic speech data. In line with the zero-resource paradigm, an ABX test was then used to investigate how the extracted representations encode phonetically relevant properties. Experiments were conducted on three different corpora in English and French. We found that articulatory information rather organises the latent representations in terms of place of articulation whereas the speech acoustics mainly structure the latent space in terms of manner of articulation. We show that an optimal fusion of the two modalities can lead to a joint representation of these phonetic dimensions more accurate than each modality considered individually. Since articulatory information is usually not available in a practical situation, we finally investigate the benefit it provides when inferred from the speech acoustics in a self-supervised manner.

Multistream neural architectures for cued-speech recognition using a pre-trained visual feature extractor and constrained CTC decoding

This paper proposes a simple and effective approach for automatic recognition of Cued Speech (CS), a visual communication tool that helps people with hearing impairment to understand spoken language with the help of hand gestures that can uniquely identify the uttered phonemes in complement to lipreading. The proposed approach is based on a pre-trained hand and lips tracker used for visual feature extraction and a phonetic decoder based on a multistream recurrent neural network trained with connectionist temporal classification loss and combined with a pronunciation lexicon. The proposed system is evaluated on an updated version of the French CS dataset CSF18 for which the phonetic transcription has been manually checked and corrected. With a decoding accuracy at the phonetic level of 70.88%, the proposed system outperforms our previous CNN-HMM decoder and competes with more complex baselines.

Repeat after me: Self-supervised learning of acoustic-to-articulatory mapping by vocal imitation

We propose a computational model of speech production combining a pre-trained neural articulatory synthesizer able to reproduce complex speech stimuli from a limited set of interpretable articulatory parameters, a DNN-based internal forward model predicting the sensory consequences of articulatory commands, and an internal inverse model based on a recurrent neural network recovering articulatory commands from the acoustic speech input. Both forward and inverse models are jointly trained in a self-supervised way from raw acoustic-only speech data from different speakers. The imitation simulations are evaluated objectively and subjectively and display quite encouraging performances.

A Benchmark of Dynamical Variational Autoencoders applied to Speech Spectrogram Modeling

The Variational Autoencoder (VAE) is a powerful deep generative model that is now extensively used to represent high-dimensional complex data via a low-dimensional latent space learned in an unsupervised manner. In the original VAE model, input data vectors are processed independently. In recent years, a series of papers have presented different extensions of the VAE to process sequential data, that not only model the latent space, but also model the temporal dependencies within a sequence of data vectors and corresponding latent vectors, relying on recurrent neural networks. We recently performed a comprehensive review of those models and unified them into a general class called Dynamical Variational Autoencoders (DVAEs). In the present paper, we present the results of an experimental benchmark comparing six of those DVAE models on the speech analysis-resynthesis task, as an illustration of the high potential of DVAEs for speech modeling.

Learning robust speech representation with an articulatory-regularized variational autoencoder

It is increasingly considered that human speech perception and production both rely on articulatory representations. In this paper, we investigate whether this type of representation could improve the performances of a deep generative model (here a variational autoencoder) trained to encode and decode acoustic speech features. First we develop an articulatory model able to associate articulatory parameters describing the jaw, tongue, lips and velum configurations with vocal tract shapes and spectral features. Then we incorporate these articulatory parameters into a variational autoencoder applied on spectral features by using a regularization technique that constraints part of the latent space to follow articulatory trajectories. We show that this articulatory constraint improves model training by decreasing time to convergence and reconstruction loss at convergence, and yields better performance in a speech denoising task.

Alternate Endings: Improving Prosody for Incremental Neural TTS with Predicted Future Text Input

The prosody of a spoken word is determined by its surrounding context. In incremental text-to-speech synthesis, where the synthesizer produces an output before it has access to the complete input, the full context is often unknown which can result in a loss of naturalness in the synthesized speech. In this paper, we investigate whether the use of predicted future text can attenuate this loss. We compare several test conditions of next future word: (a) unknown (zero-word), (b) language model predicted, (c) randomly predicted and (d) ground-truth. We measure the prosodic features (pitch, energy and duration) and find that predicted text provides significant improvements over a zero-word lookahead, but only slight gains over random-word lookahead. We confirm these results with a perceptive test.