# Hard-Synth: Synthesizing Diverse Hard Samples for ASR using Zero-Shot TTS and LLM

arXiv: https://arxiv.org/abs/2411.13159 · HTML: https://mlsanity.com/paper/2411.13159/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 37 references that are in the map.
- First posted on arXiv 20 Nov 2024.

## Builds on (most cited first, 30 of 30)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition](https://mlsanity.com/paper/1904.08779/) (2019)
- [GLM-130B: An Open Bilingual Pre-trained Model](https://mlsanity.com/paper/2210.02414/) (2022)
- [Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers](https://mlsanity.com/paper/2301.02111/) (2023)
- [ESPnet: End-to-End Speech Processing Toolkit](https://mlsanity.com/paper/1804.00015/) (2018)
- [Attention-Based Models for Speech Recognition](https://mlsanity.com/paper/1506.07503/) (2015)
- [F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching](https://mlsanity.com/paper/2410.06885/) (2024)
- [NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion Models](https://mlsanity.com/paper/2403.03100/) (2024)
- [MOSNet: Deep Learning based Objective Assessment for Voice Conversion](https://mlsanity.com/paper/1904.08352/) (2019)
- [VoiceCraft: Zero-Shot Speech Editing and Text-to-Speech in the Wild](https://mlsanity.com/paper/2403.16973/) (2024)
- [RewriteLM: An Instruction-Tuned Large Language Model for Text Rewriting](https://mlsanity.com/paper/2305.15685/) (2023)
- [Can Generative Large Language Models Perform ASR Error Correction?](https://mlsanity.com/paper/2307.04172/) (2023)
- [Quantifying Bias in Automatic Speech Recognition](https://mlsanity.com/paper/2103.15122/) (2021)
- [Using Synthetic Audio to Improve The Recognition of Out-Of-Vocabulary Words in End-To-End ASR Systems](https://mlsanity.com/paper/2011.11564/) (2020)
- [Synt++: Utilizing Imperfect Synthetic Data to Improve Speech Recognition](https://mlsanity.com/paper/2110.11479/) (2021)
- [As Good As A Coin Toss: Human detection of AI-generated images, videos, audio, and audiovisual stimuli](https://mlsanity.com/paper/2403.16760/) (2024)
- [Human Perception of Audio Deepfakes](https://mlsanity.com/paper/2107.09667/) (2021)
- [Task Arithmetic can Mitigate Synthetic-to-Real Gap in Automatic Speech Recognition](https://mlsanity.com/paper/2406.02925/) (2024)
- [Enhancing Low-Resource ASR through Versatile TTS: Bridging the Data Gap](https://mlsanity.com/paper/2410.16726/) (2024)
- [Text-only domain adaptation for end-to-end ASR using integrated text-to-mel-spectrogram generator](https://mlsanity.com/paper/2302.14036/) (2023)
- [Synthesizing Dysarthric Speech Using Multi-talker TTS for Dysarthric Speech Recognition](https://mlsanity.com/paper/2201.11571/) (2022)
- [Text is All You Need: Personalizing ASR Models using Controllable Speech Synthesis](https://mlsanity.com/paper/2303.14885/) (2023)
- [Generating Data with Text-to-Speech and Large-Language Models for Conversational Speech Recognition](https://mlsanity.com/paper/2408.09215/) (2024)
- [Corpus Synthesis for Zero-shot ASR domain Adaptation using Large Language Models](https://mlsanity.com/paper/2309.10707/) (2023)
- [Improving Code-Switching and Named Entity Recognition in ASR with Speech Editing based Data Augmentation](https://mlsanity.com/paper/2306.08588/) (2023)

## Built on by (most cited first, 1 of 1)

- [Phoneme-guided TTS augmentation for ASR: A unified pipeline and multilingual evaluation](https://mlsanity.com/paper/2608.26697/) (2026)

## Abstract (arXiv)

Text-to-speech (TTS) models have been widely adopted to enhance automatic speech recognition (ASR) systems using text-only corpora, thereby reducing the cost of labeling real speech data. Existing research primarily utilizes additional text data and predefined speech styles supported by TTS models. In this paper, we propose Hard-Synth, a novel ASR data augmentation method that leverages large language models (LLMs) and advanced zero-shot TTS. Our approach employs LLMs to generate diverse in-domain text through rewriting, without relying on additional text data. Rather than using predefined speech styles, we introduce a hard prompt selection method with zero-shot TTS to clone speech styles that the ASR model finds challenging to recognize. Experiments demonstrate that Hard-Synth significantly enhances the Conformer model, achieving relative word error rate (WER) reductions of 6.5%/4.4% on LibriSpeech dev/test-other subsets. Additionally, we show that Hard-Synth is data-efficient and capable of reducing bias in ASR.