mlsanityData to 14 Oct 2026

Text-To-Speech Data Augmentation for Low Resource Speech Recognition

arXiv 2204.00291 ↗First posted 1 Apr 2022cs.CL cs.SD eess.AScited by 6

Not on the map. Why?

Abstract (arXiv)

Nowadays, the main problem of deep learning techniques used in the development of automatic speech recognition (ASR) models is the lack of transcribed data. The goal of this research is to propose a new data augmentation method to improve ASR models for agglutinative and low-resource languages. This novel data augmentation method generates both synthetic text and synthetic audio. Some experiments were conducted using the corpus of the Quechua language, which is an agglutinative and low-resource language. In this study, a sequence-to-sequence (seq2seq) model was applied to generate synthetic text, in addition to generating synthetic speech using a text-to-speech (TTS) model for Quechua. The results show that the new data augmentation method works well to improve the ASR model for Quechua. In this research, an 8.73% improvement in the word-error-rate (WER) of the ASR model is obtained using a combination of synthetic text and synthetic speech.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 21 of its 70 references that are in the map. First posted on arXiv 1 Apr 2022.

Builds on

21 of 70 references.

References · 21

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2014Sequence to Sequence Learning with Neural Networkscited by 2,986
  3. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  4. 2020Conformer: Convolution-augmented Transformer for Speech Recognitioncited by 1,863
  5. 2019SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognitioncited by 1,380
  6. 2019EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Taskscited by 807
  7. 2017Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictionscited by 804
  8. 2015Deep Speech 2: End-to-End Speech Recognition in English and Mandarincited by 451
  9. 2014Deep Speech: Scaling up end-to-end speech recognitioncited by 354
  10. 2017OpenNMT: Open-source Toolkit for Neural Machine Translationcited by 251
  11. 2018QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehensioncited by 164
  12. 2020FastPitch: Parallel Text-to-speech with Pitch Predictioncited by 151
9 more
  1. 2015Character-Aware Neural Language Modelscited by 132
  2. 2019End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architecturescited by 120
  3. 2018Sequence-to-Sequence Data Augmentation for Dialogue Language Understandingcited by 68
  4. 2019Generating Synthetic Audio Data for Attention-Based Speech Recognition Systemscited by 47
  5. 2018wav2letter++: The Fastest Open-source Speech Recognition Systemcited by 42
  6. 2019Meta Learning for End-to-End Low-Resource Speech Recognitioncited by 29
  7. 2019Improving sequence-to-sequence speech recognition training with on-the-fly data augmentationcited by 29
  8. 2014First-Pass Large Vocabulary Continuous Speech Recognition using Bi-Directional Recurrent DNNscited by 28
  9. 2020End-to-End Automatic Speech Recognition Integrated With CTC-Based Voice Activity Detectioncited by 20

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 0; 2025: 3; 2026: 2.

6 papers.

  1. 2025SoundMind: RL-Incentivized Logic Reasoning for Audio-Language Modelscited by 15
  2. 2025Synthetic Voice Data for Automatic Speech Recognition in African Languagescited by 3
  3. 2025Protect: Towards Robust Guardrailing Stack for Trustworthy Enterprise LLM Systemscited by 3
  4. 2026Improving Code-Switching ASR with Code-Mixing Guided Synthetic Speechcited by 1
  5. 2023Deepfake audio as a data augmentation technique for training automatic speech to text transcription modelscited by 0
  6. 2026Diarization-Guided Qwen-ASR Adaptation for Multilingual Two-Speaker Conversational Speechcited by 0