mlsanityData to 7 Oct 2026

Fast DCTTS: Efficient Deep Convolutional Text-to-Speech

arXiv 2104.00624 ↗First posted 1 Apr 2021eess.AS cs.AI cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

We propose an end-to-end speech synthesizer, Fast DCTTS, that synthesizes speech in real time on a single CPU thread. The proposed model is composed of a carefully-tuned lightweight network designed by applying multiple network reduction and fidelity improvement techniques. In addition, we propose a novel group highway activation that can compromise between computational efficiency and the regularization effect of the gating mechanism. As well, we introduce a new metric called Elastic mel-cepstral distortion (EMCD) to measure the fidelity of the output mel-spectrogram. In experiments, we analyze the effect of the acceleration techniques on speed and speech quality. Compared with the baseline model, the proposed model exhibits improved MOS from 2.62 to 2.74 with only 1.76% computation and 2.75% parameters. The speed on a single CPU thread was improved by 7.45 times, which is fast enough to produce mel-spectrogram in real time without GPU.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 18 of its 25 references that are in the map. First posted on arXiv 1 Apr 2021.

Builds on

18 of 25 references.

References · 18

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  4. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  5. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
  6. 2017ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devicescited by 1,194
  7. 2016Pruning Filters for Efficient ConvNetscited by 1,129
  8. 2016SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model sizecited by 1,054
  9. 2017Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictionscited by 800
  10. 2015Scheduled Sampling for Sequence Prediction with Recurrent Neural Networkscited by 779
  11. 2020FastSpeech 2: Fast and High-Quality End-to-End Text to Speechcited by 710
  12. 2016Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networkscited by 541
6 more
  1. 2017Tacotron: Towards End-to-End Speech Synthesiscited by 487
  2. 2019Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogramcited by 337
  3. 2015Highway Networkscited by 257
  4. 2017Efficiently Trainable Text-to-Speech System Based on Deep Convolutional Networks with Guided Attentioncited by 57
  5. 2020AlignTTS: Efficient Feed-Forward Text-to-Speech System without Explicit Alignmentcited by 20
  6. 2019Effective Use of Variational Embedding Capacity in Expressive End-to-End Speech Synthesiscited by 15

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 1; 2023: 1; 2024: 1.

4 papers.

  1. 2021A Survey on Neural Speech Synthesiscited by 182
  2. 2023LightGrad: Lightweight Diffusion Probabilistic Model for Text-to-Speechcited by 13
  3. 2024Lightweight Zero-shot Text-to-Speech with Mixture of Adapterscited by 2
  4. 2022Karaoker: Alignment-free singing voice synthesis with speech training datacited by 1