Generating Synthetic Audio Data for Attention-Based Speech Recognition Systems

IEEE International Conference on Acoustics Speech and Signal Processing, pp. 7069–7073

Abstract

Recent advances in text-to-speech (TTS) led to the development of flexible multi-speaker end-to-end TTS systems. We extend state-of-the-art attention-based automatic speech recognition (ASR) systems with synthetic audio generated by a TTS system trained only on the ASR corpora itself. ASR and TTS systems are built separately to show that text-only data can be used to enhance existing end-to-end ASR systems without the necessity of parameter or architecture changes. We compare our method with language model integration of the same text data and with simple data augmentation methods like SpecAugment and show that performance improvements are mostly independent. We achieve improvements of up to 33% relative in word-error-rate (WER) over a strong baseline with data-augmentation in a low-resource environment (LibriSpeech-100h), closing the gap to a comparable oracle experiment by more than 50%. We also show improvements of up to 5% relative WER over our most recent ASR baseline on LibriSpeech-960h.

Authors 4

1. Nick Rossenbach Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Human Language Technology and Pattern Recognition,Germany

   RWTH Aachen University

2. Albert Zeyer Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Human Language Technology and Pattern Recognition,Germany

3. Ralf Schlüter Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Human Language Technology and Pattern Recognition,Germany

4. Hermann Ney Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University,Human Language Technology and Pattern Recognition,Germany