NatiQ: An End-to-end Text-to-Speech System for Arabic
Preprint
Abstract
NatiQ is end-to-end text-to-speech system for Arabic. Our speech synthesizer uses an encoder-decoder architecture with attention. We used both tacotron-based models (tacotron-1 and tacotron-2) and the faster transformer model for generating mel-spectrograms from characters. We concatenated Tacotron1 with the WaveRNN vocoder, Tacotron2 with the WaveGlow vocoder and ESPnet transformer with the parallel wavegan vocoder to synthesize waveforms from the spectrograms. We used in-house speech data for two voices: 1) neutral male "Hamza"- narrating general content and news, and 2) expressive female "Amina"- narrating children story books to train our models. Our best systems achieve an average Mean Opinion Score (MOS) of 4.21 and 4.40 for Amina and Hamza respectively. The objective evaluation of the systems using word and character error rate (WER and CER) as well as the response time measured by real-time factor favored the end-to-end architecture ESPnet. NatiQ demo is available on-line at https://tts.qcri.orgKeywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Arabic speech recognition by end-to-end, modular systems and humanComputer Speech & Language, 2022
- Expressive TTS Training With Frame and Style Reconstruction LossIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
- Parallel Wavegan: A Fast Waveform Generation Model Based on Generative Adversarial Networks with Multi-Resolution Spectrogram2020
- Mellotron: Multispeaker Expressive Voice Synthesis by Conditioning on Rhythm, Pitch and Global Style Tokens2020
- The MGB-5 Challenge: Recognition and Dialect Identification of Dialectal Arabic Speech2019
- Waveglow: A Flow-based Generative Network for Speech Synthesis2019
- Forward Attention in Sequence- To-Sequence Acoustic Modeling for Speech Synthesis2018
- Natural TTS Synthesis by Conditioning Wavenet on MEL Spectrogram Predictions2018
- Speech recognition challenge in the wild: Arabic MGB-32017
- Perception of Expressivity in TTS: Linguistics, Phonetics or Prosody?2017