Improving Opus Low Bit Rate Quality with Neural Speech Synthesis

@inproceedings{Skoglund2020ImprovingOL,
  title={Improving Opus Low Bit Rate Quality with Neural Speech Synthesis},
  author={J. Skoglund and J. Valin},
  booktitle={INTERSPEECH},
  year={2020}
}
The voice mode of the Opus audio coder can compress wideband speech at bit rates ranging from 6 kb/s to 40 kb/s. However, Opus is at its core a waveform matching coder, and as the rate drops below 10 kb/s, quality degrades quickly. As the rate reduces even further, parametric coders tend to perform better than waveform coders. In this paper we propose a backward-compatible way of improving low bit rate Opus quality by re-synthesizing speech from the decoded parameters. We compare two different… Expand
6 Citations
Enhancement of Coded Speech Using a Mask-Based Post-Filter
  • PDF
Speech Quality Factors for Traditional and Neural-Based Low Bit Rate Vocoders
  • 1
  • PDF
Audio Codec Enhancement with Generative Adversarial Networks
  • A. Biswas, Dai Jia
  • Computer Science, Engineering
  • ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2020
  • 4
  • PDF
Source Coding of Audio Signals with a Generative Model
  • 2
  • PDF
Speech Resynthesis from Discrete Disentangled Self-Supervised Representations
  • PDF

References

SHOWING 1-10 OF 33 REFERENCES
Low Bit-rate Speech Coding with VQ-VAE and a WaveNet Decoder
  • 23
  • PDF
Wavenet Based Low Rate Speech Coding
  • 51
  • PDF
A Real-Time Wideband Neural Vocoder at 1.6 kb/s Using LPCNet
  • 26
  • PDF
Code-excited linear prediction(CELP): High-quality speech at very low bit rates
  • M. Schroeder, B. Atal
  • Computer Science
  • ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing
  • 1985
  • 1,104
  • Highly Influential
  • PDF
Adaptive noise spectral shaping and entropy coding in predictive coding of speech
  • 82
  • Highly Influential
High-quality Speech Coding with Sample RNN
  • 17
  • PDF
Efficient Neural Audio Synthesis
  • 357
  • Highly Influential
  • PDF
LPCNET: Improving Neural Speech Synthesis through Linear Prediction
  • J. Valin, J. Skoglund
  • Computer Science, Engineering
  • ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2019
  • 151
  • PDF
Voice Coding with Opus
  • 26
  • PDF
A 2.4 kbit/s MELP coder candidate for the new U.S. Federal Standard
  • 162
...
1
2
3
4
...