# Speech waveform synthesis from MFCC sequences with generative adversarial networks

arXiv: https://arxiv.org/abs/1804.00920 · HTML: https://mlsanity.com/paper/1804.00920/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Apr 2018.

## Built on by (most cited first, 5 of 5)

- [LongVALE: Vision-Audio-Language-Event Benchmark Towards Time-Aware Omni-Modal Perception of Long Videos](https://mlsanity.com/paper/2411.19772/) (2024)
- [Diff-TTSG: Denoising probabilistic integrated speech and gesture synthesis](https://mlsanity.com/paper/2306.09417/) (2023)
- [DM-Codec: Distilling Multimodal Representations for Speech Tokenization](https://mlsanity.com/paper/2410.15017/) (2024)
- [Wavebender GAN: An architecture for phonetically meaningful speech manipulation](https://mlsanity.com/paper/2202.10973/) (2022)
- [MFCCGAN: A Novel MFCC-Based Speech Synthesizer Using Adversarial Learning](https://mlsanity.com/paper/2306.12785/) (2023)

## Abstract (arXiv)

This paper proposes a method for generating speech from filterbank mel frequency cepstral coefficients (MFCC), which are widely used in speech applications, such as ASR, but are generally considered unusable for speech synthesis. First, we predict fundamental frequency and voicing information from MFCCs with an autoregressive recurrent neural net. Second, the spectral envelope information contained in MFCCs is converted to all-pole filters, and a pitch-synchronous excitation model matched to these filters is trained. Finally, we introduce a generative adversarial network -based noise model to add a realistic high-frequency stochastic component to the modeled excitation signal. The results show that high quality speech reconstruction can be obtained, given only MFCC information at test time.