A Study on Incorporating Whisper for Robust Speech Assessment

This research introduces an enhanced version of the multi-objective speech assessment model, called MOSA-Net+, by leveraging the acoustic features from large pre-trained weakly supervised models, namely Whisper, to create embedding features. The first part of this study investigates the correlation between the embedding features of Whisper and two self-supervised learning (SSL) models with subjective quality and intelligibility scores. The second part evaluates the effectiveness of Whisper in deploying a more robust speech assessment model. Third, the possibility of combining representations from Whisper and SSL models while deploying MOSA-Net+ is analyzed. The experimental results reveal that Whisper's embedding features correlate more strongly with subjective quality and intelligibility than other SSL's embedding features, contributing to more accurate prediction performance achieved by MOSA-Net+. Moreover, combining the embedding features from Whisper and SSL models only leads to marginal improvement. As compared to MOSA-Net and other SSL-based speech assessment models, MOSA-Net+ yields notable improvements in estimating subjective quality and intelligibility scores across all evaluation metrics. We further tested MOSA-Net+ on Track 3 of the VoiceMOS Challenge 2023 and obtained the top-ranked performance.

Noise robust speech emotion recognition with signal-to-noise ratio adapting speech enhancement

Speech emotion recognition (SER) often experiences reduced performance due to background noise. In addition, making a prediction on signals with only background noise could undermine user trust in the system. In this study, we propose a Noise Robust Speech Emotion Recognition system, NRSER. NRSER employs speech enhancement (SE) to effectively reduce the noise in input signals. Then, the signal-to-noise-ratio (SNR)-level detection structure and waveform reconstitution strategy are introduced to reduce the negative impact of SE on speech signals with no or little background noise. Our experimental results show that NRSER can effectively improve the noise robustness of the SER system, including preventing the system from making emotion recognition on signals consisting solely of background noise. Moreover, the proposed SNR-level detection structure can be used individually for tasks such as data selection.

MultiPA: a multi-task speech pronunciation assessment system for a closed and open response scenario

The design of automatic speech pronunciation assessment can be categorized into closed and open response scenarios, each with strengths and limitations. A system with the ability to function in both scenarios can cater to diverse learning needs and provide a more precise and holistic assessment of pronunciation skills. In this study, we propose a Multi-task Pronunciation Assessment model called MultiPA. MultiPA provides an alternative to Kaldi-based systems in that it has simpler format requirements and better compatibility with other neural network models. Compared with previous open response systems, MultiPA provides a wider range of evaluations, encompassing assessments at both the sentence and word-level. Our experimental results show that MultiPA achieves comparable performance when working in closed response scenarios and maintains more robust performance when directly used for open responses.

Investigation of Factorized Optical Flows as Mid-Level Representations

In this paper, we introduce a new concept of incorporating factorized flow maps as mid-level representations, for bridging the perception and the control modules in modular learning based robotic frameworks. To investigate the advantages of factorized flow maps and examine their interplay with the other types of mid-level representations, we further develop a configurable framework, along with four different environments that contain both static and dynamic objects, for analyzing the impacts of factorized optical flow maps on the performance of deep reinforcement learning agents. Based on this framework, we report our experimental results on various scenarios, and offer a set of analyses to justify our hypothesis. Finally, we validate flow factorization in real world scenarios.

InQSS: a speech intelligibility assessment model using a multi-task learning network

Speech intelligibility assessment models are essential tools for researchers to evaluate and improve speech processing models. In this study, we propose InQSS, a speech intelligibility assessment model that uses both spectrogram and scattering coefficients as input features. In addition, InQSS uses a multi-task learning network in which quality scores can guide the training of the speech intelligibility assessment. The resulting model can predict not only the intelligibility scores but also the quality scores of a speech. The experimental results confirm that the scattering coefficients and quality scores are informative for intelligibility. Moreover, we released TMHINT-QI, which is a Chinese speech dataset that records the quality and intelligibility scores of clean, noisy, and enhanced speech.

The AS-NU System for the M2VoC Challenge

This paper describes the AS-NU systems for two tracks in MultiSpeaker Multi-Style Voice Cloning Challenge (M2VoC). The first track focuses on using a small number of 100 target utterances for voice cloning, while the second track focuses on using only 5 target utterances for voice cloning. Due to the serious lack of data in the second track, we selected the speaker most similar to the target speaker from the training data of the TTS system, and used the speaker's utterances and the given 5 target utterances to fine-tune our model. The evaluation results show that our systems on the two tracks perform similarly in terms of quality, but there is still a clear gap between the similarity score of the second track and the similarity score of the first track.

EMA2S: An End-to-End Multimodal Articulatory-to-Speech System

Synthesized speech from articulatory movements can have real-world use for patients with vocal cord disorders, situations requiring silent speech, or in high-noise environments. In this work, we present EMA2S, an end-to-end multimodal articulatory-to-speech system that directly converts articulatory movements to speech signals. We use a neural-network-based vocoder combined with multimodal joint-training, incorporating spectrogram, mel-spectrogram, and deep features. The experimental results confirm that the multimodal approach of EMA2S outperforms the baseline system in terms of both objective evaluation and subjective evaluation metrics. Moreover, results demonstrate that joint mel-spectrogram and deep feature loss training can effectively improve system performance.

Investigating the Ground-level Ozone Formation and Future Trend in Taiwan

Tropospheric ozone (O3) is an influential ground-level air pollutant which can severely damage the environment. Thus evaluating the importance of various factors related to the O3 formation process is essential. However, O3 simulated by the available climate models exhibits large variance in different places, indicating the insufficiency of models in explaining the O3 formation process correctly. In this paper, we aim to understand the impact of various factors on O3 formation and predict the O3 concentrations. Six well-known supervised learning methods are evaluated to estimate the observed O3 using sixteen meteorological and chemical variables. We find that the XGBoost and the convolution neural network (CNN) models achieve most accurate predictions. We also demonstrate the importance of several variables empirically. The results suggest that while Nitrogen Oxides negatively contributes to predicting O3, the amount of solar radiation makes significantly positive contribution. Furthermore, we apply the XGBoost model on climate O3 prediction and show its competence in calibrating the O3 simulated by a global climate model.

CITISEN: A Deep Learning-Based Speech Signal-Processing Mobile Application

In this paper, we present a deep learning-based speech signal-processing mobile application, CITISEN, which can perform three functions: speech enhancement (SE), acoustic scene conversion (ASC), and model adaptation (MA). For SE, CITISEN can effectively reduce noise components from speech signals and accordingly enhance their clarity and intelligibility. For ASC, CITISEN can convert the current background sound to a different background sound. Finally, for MA, CITISEN can effectively adapt an SE model, with a few audio files, when it encounters unknown speakers or noise types; the adapted SE model is used to enhance the upcoming noisy utterances. Experimental results confirmed the effectiveness of CITISEN in performing these three functions via objective evaluation and subjective listening tests. The promising results reveal that the developed CITISEN mobile application can potentially be used as a front-end processor for various speech-related services such as voice communication, assistive hearing devices, and virtual reality headsets.