English
Related papers

Related papers: MELD: Mel-Spectrogram-Based Speech Language Modeli…

200 papers

Neural network based end-to-end text to speech (TTS) has significantly improved the quality of synthesized speech. Prominent methods (e.g., Tacotron 2) usually first generate mel-spectrogram from text, and then synthesize speech from the…

Computation and Language · Computer Science 2019-11-21 Yi Ren , Yangjun Ruan , Xu Tan , Tao Qin , Sheng Zhao , Zhou Zhao , Tie-Yan Liu

Vocoders are models capable of transforming a low-dimensional spectral representation of an audio signal, typically the mel spectrogram, to a waveform. Modern speech generation pipelines use a vocoder as their final component. Recent…

Sound · Computer Science 2022-08-29 Bruno Di Giorgi , Mark Levy , Richard Sharp

Historically, most speech models in machine-learning have used the mel-spectrogram as a speech representation. Recently, discrete audio tokens produced by neural audio codecs have become a popular alternate speech representation for speech…

Audio and Speech Processing · Electrical Eng. & Systems 2025-06-05 Ryan Langman , Ante Jukić , Kunal Dhawan , Nithin Rao Koluguri , Jason Li

Neural audio codecs (NACs) provide compact latent speech representations in the form of sequences of continuous vectors or discrete tokens. In this work, we investigate how these two types of speech representations compare when used as…

Sound · Computer Science 2026-03-12 Sofiene Kammoun , Xavier Alameda-Pineda , Simon Leglaive

In speech synthesis and speech enhancement systems, melspectrograms need to be precise in acoustic representations. However, the generated spectrograms are over-smooth, that could not produce high quality synthesized speech. Inspired by…

Audio and Speech Processing · Electrical Eng. & Systems 2019-12-04 Leyuan Sheng , Dong-Yan Huang , Evgeniy N. Pavlovskiy

A latent-variable model is introduced for text matching, inferring sentence representations by jointly optimizing generative and discriminative objectives. To alleviate typical optimization challenges in latent-variable models for text, we…

Computation and Language · Computer Science 2017-11-23 Dinghan Shen , Yizhe Zhang , Ricardo Henao , Qinliang Su , Lawrence Carin

Integrating front-end speech enhancement (SE) models with self-supervised learning (SSL)-based speech models is effective for downstream tasks in noisy conditions. SE models are commonly fine-tuned using SSL representations with mean…

Computation and Language · Computer Science 2026-01-30 Amit Meghanani , Thomas Hain

Previously, we introduced VoiceGrad, a nonparallel voice conversion (VC) technique enabling mel-spectrogram conversion from source to target speakers using a score-based diffusion model. The concept involves training a score network to…

Sound · Computer Science 2025-09-11 Hirokazu Kameoka , Takuhiro Kaneko , Kou Tanaka , Yuto Kondo

We propose a novel Multi-Scale Spectrogram (MSS) modelling approach to synthesise speech with an improved coarse and fine-grained prosody. We present a generic multi-scale spectrogram prediction mechanism where the system first predicts…

Audio and Speech Processing · Electrical Eng. & Systems 2021-07-01 Ammar Abbas , Bajibabu Bollepalli , Alexis Moinet , Arnaud Joly , Penny Karanasou , Peter Makarov , Simon Slangens , Sri Karlapati , Thomas Drugman

With latent variables, stochastic recurrent models have achieved state-of-the-art performance in modeling sound-wave sequence. However, opposite results are also observed in other domains, where standard recurrent networks often outperform…

Machine Learning · Computer Science 2019-09-17 Zihang Dai , Guokun Lai , Yiming Yang , Shinjae Yoo

Recently, neural network-based methods for computing optimal transport maps have been effectively applied to style transfer problems. However, the application of these methods to voice conversion is underexplored. In our paper, we fill this…

Variational encoder-decoders (VEDs) have shown promising results in dialogue generation. However, the latent variable distributions are usually approximated by a much simpler model than the powerful RNN structure used for encoding and…

Computation and Language · Computer Science 2018-02-07 Xiaoyu Shen , Hui Su , Shuzi Niu , Vera Demberg

Nearly all Statistical Parametric Speech Synthesizers today use Mel Cepstral coefficients as the vocal tract parameterization of the speech signal. Mel Cepstral coefficients were never intended to work in a parametric speech synthesis…

Computation and Language · Computer Science 2014-10-01 Prasanna Kumar Muthukumar , Alan W. Black

There are many deterministic mathematical operations (e.g. compression, clipping, downsampling) that degrade speech quality considerably. In this paper we introduce a neural network architecture, based on a modification of the DiffWave…

Sound · Computer Science 2021-09-03 Jianwei Zhang , Suren Jayasuriya , Visar Berisha

Text-to-speech (TTS) methods have shown promising results in voice cloning, but they require a large number of labeled text-speech pairs. Minimally-supervised speech synthesis decouples TTS by combining two types of discrete speech…

Sound · Computer Science 2023-12-19 Chunyu Qiang , Hao Li , Yixin Tian , Yi Zhao , Ying Zhang , Longbiao Wang , Jianwu Dang

We present an approach for unsupervised learning of speech representation disentangling contents and styles. Our model consists of: (1) a local encoder that captures per-frame information; (2) a global encoder that captures per-utterance…

Computation and Language · Computer Science 2021-06-22 Andros Tjandra , Ruoming Pang , Yu Zhang , Shigeki Karita

Introducing variability while maintaining coherence is a core task in learning to generate utterances in conversation. Standard neural encoder-decoder models and their extensions using conditional variational autoencoder often result in…

Computation and Language · Computer Science 2018-10-23 Hung Le , Truyen Tran , Thin Nguyen , Svetha Venkatesh

In this paper, a neural network named Sequence-to-sequence ConvErsion NeTwork (SCENT) is presented for acoustic modeling in voice conversion. At training stage, a SCENT model is estimated by aligning the feature sequences of source and…

Sound · Computer Science 2020-01-14 Jing-Xuan Zhang , Zhen-Hua Ling , Li-Juan Liu , Yuan Jiang , Li-Rong Dai

This paper proposes a novel unsupervised autoregressive neural model for learning generic speech representations. In contrast to other speech representation learning methods that aim to remove noise or speaker variabilities, ours is…

Computation and Language · Computer Science 2019-06-20 Yu-An Chung , Wei-Ning Hsu , Hao Tang , James Glass

The modern autoregressive Large Language Models (LLMs) have achieved outstanding performance on NLP benchmarks, and they are deployed in the real world. However, they still suffer from limitations of the autoregressive training paradigm.…

Computation and Language · Computer Science 2024-07-11 Justin Deschenaux , Caglar Gulcehre