English
Related papers

Related papers: Real-Time Streaming Mel Vocoding with Generative F…

200 papers

Diffusion-based audio and music generation models commonly perform generation by constructing an image representation of audio (e.g., a mel-spectrogram) and then convert it to audio using a phase reconstruction model or vocoder. Typical…

Sound · Computer Science 2024-10-08 Ge Zhu , Juan-Pablo Caceres , Zhiyao Duan , Nicholas J. Bryan

The latency bottleneck of traditional text-to-speech (TTS) systems fundamentally hinders the potential of streaming large language models (LLMs) in conversational AI. These TTS systems, typically trained and inferenced on complete…

Computation and Language · Computer Science 2025-05-27 Richard He Bai , Zijin Gu , Tatiana Likhomanenko , Navdeep Jaitly

Recently, GAN vocoders have seen rapid progress in speech synthesis, starting to outperform autoregressive models in perceptual quality with much higher generation speed. However, autoregressive vocoders are still the common choice for…

Audio and Speech Processing · Electrical Eng. & Systems 2021-08-10 Ahmed Mustafa , Jan Büthe , Srikanth Korse , Kishan Gupta , Guillaume Fuchs , Nicola Pia

We present a novel high-fidelity real-time neural vocoder called VocGAN. A recently developed GAN-based vocoder, MelGAN, produces speech waveforms in real-time. However, it often produces a waveform that is insufficient in quality or…

Audio and Speech Processing · Electrical Eng. & Systems 2020-07-31 Jinhyeok Yang , Junmo Lee , Youngik Kim , Hoonyoung Cho , Injung Kim

Transformer-based models have achieved state-of-the-art performance on speech translation tasks. However, the model architecture is not efficient enough for streaming scenarios since self-attention is computed over an entire input sequence…

Computation and Language · Computer Science 2020-11-03 Xutai Ma , Yongqiang Wang , Mohammad Javad Dousti , Philipp Koehn , Juan Pino

Typical high quality text-to-speech (TTS) systems today use a two-stage architecture, with a spectrum model stage that generates spectral frames and a vocoder stage that generates the actual audio. High-quality spectrum models usually…

Sound · Computer Science 2021-04-05 Qing He , Zhiping Xiu , Thilo Koehler , Jilong Wu

Zero-shot streaming text-to-speech is an important research topic in human-computer interaction. Existing methods primarily use a lookahead mechanism, relying on future text to achieve natural streaming speech synthesis, which introduces…

Machine Learning · Computer Science 2025-06-03 Haiyang Sun , Shujie Hu , Shujie Liu , Lingwei Meng , Hui Wang , Bing Han , Yifan Yang , Yanqing Liu , Sheng Zhao , Yan Lu , Yanmin Qian

This paper integrates a classic mel-cepstral synthesis filter into a modern neural speech synthesis system towards end-to-end controllable speech synthesis. Since the mel-cepstral synthesis filter is explicitly embedded in neural waveform…

Audio and Speech Processing · Electrical Eng. & Systems 2022-11-22 Takenori Yoshimura , Shinji Takaki , Kazuhiro Nakamura , Keiichiro Oura , Yukiya Hono , Kei Hashimoto , Yoshihiko Nankaku , Keiichi Tokuda

This paper introduces R-MelNet, a two-part autoregressive architecture with a frontend based on the first tier of MelNet and a backend WaveRNN-style audio decoder for neural text-to-speech synthesis. Taking as input a mixed sequence of…

Sound · Computer Science 2022-07-01 Kyle Kastner , Aaron Courville

Current two-stage TTS framework typically integrates an acoustic model with a vocoder -- the acoustic model predicts a low resolution intermediate representation such as Mel-spectrum while the vocoder generates waveform from the…

Audio and Speech Processing · Electrical Eng. & Systems 2021-06-23 Jian Cong , Shan Yang , Lei Xie , Dan Su

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

Target speaker extraction (TSE) aims to isolate a desired speaker's voice from a multi-speaker mixture using auxiliary information such as a reference utterance. Although recent advances in diffusion and flow-matching models have improved…

Audio and Speech Processing · Electrical Eng. & Systems 2025-12-23 Riki Shimizu , Xilin Jiang , Nima Mesgarani

Although neural text-to-speech (TTS) models have attracted a lot of attention and succeeded in generating human-like speech, there is still room for improvements to its naturalness and architectural efficiency. In this work, we propose a…

Audio and Speech Processing · Electrical Eng. & Systems 2021-04-06 Myeonghun Jeong , Hyeongju Kim , Sung Jun Cheon , Byoung Jin Choi , Nam Soo Kim

Extracting features from the speech is the most critical process in speech signal processing. Mel Frequency Cepstral Coefficients (MFCC) are the most widely used features in the majority of the speaker and speech recognition applications,…

Sound · Computer Science 2025-10-31 Rinku Sebastian , Simon O'Keefe , Martin Trefzer

Normally, a system that translates speech into text consists of separate modules for speech recognition and text-to-text translation. Combining those tasks into a SpeechLLM promises to exploit paralinguistic information in the speech and to…

Computation and Language · Computer Science 2026-05-15 Titouan Parcollet , Shucong Zhang , Xianrui Zheng , Rogier C. van Dalen

In this paper we propose WaveGlow: a flow-based network capable of generating high quality speech from mel-spectrograms. WaveGlow combines insights from Glow and WaveNet in order to provide fast, efficient and high-quality audio synthesis,…

Sound · Computer Science 2018-11-02 Ryan Prenger , Rafael Valle , Bryan Catanzaro

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

In this paper we propose Flowtron: an autoregressive flow-based generative network for text-to-speech synthesis with control over speech variation and style transfer. Flowtron borrows insights from IAF and revamps Tacotron in order to…

Sound · Computer Science 2020-07-17 Rafael Valle , Kevin Shih , Ryan Prenger , Bryan Catanzaro

Most modern text-to-speech architectures use a WaveNet vocoder for synthesizing high-fidelity waveform audio, but there have been limitations, such as high inference time, in its practical application due to its ancestral sampling scheme.…

Sound · Computer Science 2019-05-21 Sungwon Kim , Sang-gil Lee , Jongyoon Song , Jaehyeon Kim , Sungroh Yoon

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