RUSLAN is a large Russian speech corpus for text-to-speech.
problem Lack of high-quality annotated Russian speech data for text-to-speech.
method Developed a large annotated Russian speech corpus and trained a neural network for text-to-speech synthesis.
result Synthesized speech quality evaluated with MOS scores: 4.05 for naturalness, 3.78 for intelligibility.
Speech emotion recognition system using features and text.
problem Improving accuracy in emotion recognition from speech.
method Used speech features (Spectrogram, MFCC) and text, trained Deep Neural Networks.
result Combined MFCC-Text CNN model achieved highest accuracy.
AudioPaLM combines text and speech models to improve speech processing and translation.
problem Improving speech understanding and generation models.
method Unified multimodal architecture combining text and speech models.
result AudioPaLM significantly outperforms existing speech translation systems.
Enhanced text-to-speech synthesizes expressive speech from a single example.
problem Creating a new expressive speech style from a single example of speech.
method Combines VAE and Normalizing Flows to improve disentanglement and naturalness.
result Reduces KL-divergence by 22% and improves perceptual metrics.
A new TTS method uses diffusion and VAE for better speech synthesis.
problem Improving text-to-speech synthesis for better speech quality and robustness.
method Combines diffusion probabilistic model and variational autoencoder for latent variable conversion.
result The method is robust to poor orthography and alignment errors.
Grad-TTS models speech from text using diffusion probabilistic techniques.
problem Creating high-quality speech from text input.
method Score-based decoder with stochastic differential equations for noise-to-speech transformation.
result Grad-TTS produces mel-spectrograms from text input with competitive quality.
New model predicts hate speech with uncertainty estimates.
problem Detect and explain hate speech without restricting free speech.
method Adapted deep neural networks with Monte Carlo dropout.
result Models can reliably assess when texts are borderline.
Improves text-to-speech speed by interleaving character reading and audio synthesis.
problem Latency in text-to-speech models limits their use in time-sensitive tasks.
method Reinforcement learning to train an agent to choose the order of character reading and audio synthesis.
result The proposed method successfully balances latency and audio quality.
Speech Translation has always been about giving source text or audio input and waiting for system to give translated output in desired form. In this paper, we present the Acoustic Dialect Decoder (ADD) - a voice to voice ear-piece translation device. We introduce and survey the recent advances made in the field of Spee…
Telephonetic improves robustness of neural language models to ASR errors.
problem Handling ASR errors and semantic noise in neural language models trained on written text.
method Data augmentation framework using character-level and embedding space perturbations.
result Achieves state-of-the-art perplexity of 37.49 on Penn Treebank corpus.
TBIP uses texts to quantify lawmakers' political positions.
problem Quantifying lawmakers' political positions from speeches, tweets, etc.
method Unsupervised probabilistic topic model analyzing texts.
result TBIP separates lawmakers by party and infers ideal points close to vote-based.
Deep learning predicts mental disorders from audio and text samples.
problem Predicting mental disorders from speech samples.
method Multimodal deep learning structure using various pre-trained models for audio and text embeddings, transfer learning, and auxiliary corpora.
result Acceptable accuracy in predicting mental disorders through multimodal analysis.
This paper proposes a new architecture for speaker adaptation of multi-speaker neural-network speech synthesis systems, in which an unseen speaker's voice can be built using a relatively small amount of speech data without transcriptions. This is sometimes called "unsupervised speaker adaptation". More specifically, we…
Joint training model for TTS and VC tasks using Tacotron and WaveNet.
problem Training a shared model for text-to-speech and voice conversion.
method Extended Tacotron model with dual attention mechanism for shared tasks, WaveNet for waveform generation.
result Joint training of a shared model achieves both TTS and VC tasks efficiently.
Combining data from multiple speakers improves neural TTS quality, especially with imbalanced data.
problem Training high-quality TTS systems with imbalanced speaker data.
method Combine data from multiple speakers, train multi-speaker models, and use ensemble methods.
result Ensemble multi-speaker models improve synthetic speech quality for underrepresented speakers.
Improves naturalness in TTS samples using quantized VAE and auto-regressive prosody.
problem Discontinuous and unnatural speech from standard VAE priors.
method Discretized latent features using vector quantization (VQ), and separately trained autoregressive (AR) prior model.
result Significantly improves naturalness in random sample generation.
Enhances speech emotion recognition by integrating visual data with attention mechanisms.
problem Improving emotion detection accuracy by combining multiple modalities.
method Introducing an attention mechanism to combine audio, text, and video modalities.
result Significant improvement of 3.65% in weighted accuracy.
Global Style Tokens (GSTs) are a recently-proposed method to learn latent disentangled representations of high-dimensional data. GSTs can be used within Tacotron, a state-of-the-art end-to-end text-to-speech synthesis system, to uncover expressive factors of variation in speaking style. In this work, we introduce the T…
Acoustic Neighbor Embeddings map speech and text to fixed dimensions for phonetic confusability.
problem Mapping speech and text to fixed dimensions for phonetic confusability.
method Adapting SNE to sequential inputs, training two encoder neural networks.
result More accurate results with low-dimensional embeddings in word recognition tasks.
Continuous speech recognition from brain activity without vocalization.
problem Recognizing silent speech from EEG signals.
method Implemented a CTC ASR model using EEG signals.
result Demonstrated feasibility of EEG for continuous silent speech recognition.
End-to-end speech synthesis is a promising approach that directly converts raw text to speech. Although it was shown that Tacotron2 outperforms classical pipeline systems with regards to naturalness in English, its applicability to other languages is still unknown. Japanese could be one of the most difficult languages …
We present a recurrent encoder-decoder deep neural network architecture that directly translates speech in one language into text in another. The model does not explicitly transcribe the speech into text in the source language, nor does it require supervision from the ground truth source language transcription during t…
Proposes Textual Echo Cancellation to improve speech recognition.
problem Improving speech recognition performance and user experience for smart devices.
method A novel sequence-to-sequence model with multi-source attention that processes both the microphone mixture signal and source text of TTS playback.
result Demonstrates enhanced speech recognition performance and reduced latency.
Neural network converts speech from one language to multiple languages.
problem Creating speech in multiple languages without parallel data.
method Polyglot neural network with multiple per-language sub-networks and loss terms to preserve speaker identity.
result Convincing conversion capabilities demonstrated across three languages.
AeGAN improves speech clarity in noisy environments.
problem Improving speech recognition in crowded noisy environments.
method Generative adversarial networks (GAN) with a novel architecture.
result The proposed framework outperforms traditional and learning-based methods.
Investigates neural TTS systems for Japanese and English.
problem Improving neural TTS systems for high-quality speech synthesis.
method Comparative study of neural sequence-to-sequence TTS vs. DNN pipeline TTS, varying model architecture, parameter size, and language.
result A neural sequence-to-sequence TTS system requires sufficient model parameters and a powerful encoder for high-quality speech synthesis.
ResNet with Focal Loss improves speech emotion recognition.
problem Speech emotion recognition using plain text features is insufficient.
method Residual Convolutional Neural Network (ResNet) trained with Focal Loss.
result Focal Loss enhances model's focus on hard examples.
Model learns association between text and speech without paired data.
problem Learning association between text and speech with limited paired data.
method Noisy channel generative model with variational inference and KL encoder loss.
result Tiny paired data (5 minutes) sufficient to learn association.
RNNs are crucial for text and speech tasks, explained in this overview.
problem Understanding RNNs for advanced AI tasks.
method Overview of RNN concepts and recent advances.
result Fundamentals and recent techniques in RNNs.
Guided Flows enhance sample quality in conditional image generation and text-to-speech.
problem Improving sample quality in conditional generative models.
method Integrating classifier-free guidance into Flow Matching (FM) models for Continuous Normalizing Flows (CNFs).
result Guided Flows significantly improve sample quality in conditional image generation and text-to-speech synthesis.
End-to-end model detects articulatory features from speech data.
problem Detecting articulatory features from speech data for various applications.
method Apply Listen, Attend and Spell (LAS) architecture and attention models.
result End-to-end training of manners and places of articulation detectors.
ICP improves text infilling and POS tagging with valid confidence sets.
problem Statistical reliability of machine learning predictions.
method Inductive conformal prediction algorithms for text infilling and POS tagging.
result Valid set-valued predictions with small size for real-world applications.
The state-of-the-art in text-to-speech synthesis has recently improved considerably due to novel neural waveform generation methods, such as WaveNet. However, these methods suffer from their slow sequential inference process, while their parallel versions are difficult to train and even more expensive computationally. …
Lipper synthesizes speech from silent videos, improving over single-view methods.
problem Lipreading as text classification is limited; multi-view approach needed.
method Multi-view lipreading as a regression task, producing speech from silent videos.
result Improvement in speech reconstruction with multi-view silent videos.
Study shows adversarial attacks can fool speech-to-text models, and PCA is ineffective as a defense.
problem Adversarial attacks can mislead speech-to-text neural networks.
method Crafted adversarial waveforms, used PCA for defense, tested under black-box setting.
result PCA is ineffective as a defense mechanism against adversarial attacks in audio domain.
End-to-end speaker verification framework reduces text dependency.
problem Improving text-independent speaker verification.
method Jointly trains SE and ASR networks with triplet loss and adversarial gradient.
result Lower equal error rate and better text-independency compared to other approaches.
Developers of text-to-speech synthesizers (TTS) often make use of human raters to assess the quality of synthesized speech. We demonstrate that we can model human raters' mean opinion scores (MOS) of synthesized speech using a deep recurrent neural network whose inputs consist solely of a raw waveform. Our best models …
Survey reviews code-switched speech and language processing.
problem Processing code-switched text and speech for multilingual communities.
method Reviews computational approaches and lists available resources.
result Essential for building intelligent agents that interact in multilingual settings.
Meta-learning approach discovers task-specific modules for few-shot tasks.
problem Meta-learning large models with limited task-specific data.
method Bayesian shrinkage for automatic discovery of task-specific and general modules.
result Outperforms existing meta-learning methods in domains with little task data and long adaptation horizons.
Bayesian methods improve text annotation quality.
problem Inconsistent and unreliable human annotations in natural language processing.
method Two semi-supervised Bayesian methods: a deep learning model and an ensemble method.
result Bayesian methods enhance the reliability and performance of BERT models.
Improved speech emotion recognition using pre-trained language models.
problem Challenging task of speech emotion recognition for natural human-machine interaction.
method Fine-tuning pre-trained language models for text emotion recognition, combining with speech emotion recognition.
result 73.5% accuracy in speech emotion recognition on a subset of IEMOCAP dataset.
End-to-end Text-to-speech (TTS) system can greatly improve the quality of synthesised speech. But it usually suffers form high time latency due to its auto-regressive structure. And the synthesised speech may also suffer from some error modes, e.g. repeated words, mispronunciations, and skipped words. In this paper, we…
This research investigates the Statistical Machine Translation approaches to translate speech in real time automatically. Such systems can be used in a pipeline with speech recognition and synthesis software in order to produce a real-time voice communication system between foreigners. We obtained three main data sets …
Brain2Char decodes text from brain recordings, achieving state-of-the-art performance.
problem Directly decoding text from brain recordings for communication.
method Combines 3D Inception layers, bidirectional recurrent layers, and language model weighted beam search. Uses CTC loss and auxiliary losses for regularization.
result Achieves 10.6%, 8.5%, and 7.0% WER on vocabulary sizes from 1200 to 1900 words.
Recent neural networks such as WaveNet and sampleRNN that learn directly from speech waveform samples have achieved very high-quality synthetic speech in terms of both naturalness and speaker similarity even in multi-speaker text-to-speech synthesis systems. Such neural networks are being used as an alternative to voco…
Can textual data be compressed intelligently without losing accuracy in evaluating sentiment? In this study, we propose a novel evolutionary compression algorithm, PARSEC (PARts-of-Speech for sEntiment Compression), which makes use of Parts-of-Speech tags to compress text in a way that sacrifices minimal classification…
Model predicts active and passive cosponsorship motivations in Congress.
problem Identifying motivations behind cosponsorship in U.S. Congress.
method Encoder+RGCN model learning from bill texts and speeches.
result F1-score of 0.88 for predicting active and passive cosponsorship.
Improved TTS style transfer across disjoint datasets with adversarial cycle consistency.
problem Suboptimal TTS style transfer on disjoint datasets with underrepresented styles.
method Adversarial cycle consistency training with paired and unpaired triplets.
result 78% improvement in style transfer with minimal reduction in fidelity and naturalness.