Improving Domain-Specific ASR with LLM-Generated Contextual Descriptions

• URL: http://arxiv.org/abs/2407.17874v1
• Date: Thu, 25 Jul 2024 08:44:04 GMT
• Title: Improving Domain-Specific ASR with LLM-Generated Contextual Descriptions
• Authors: Jiwon Suh, Injae Na, Woohwan Jung,
• Abstract summary: We propose a method to utilize the state-of-the-art Whisper without modifying its architecture. We also propose two additional training techniques to improve the domain specific ASR. Our experiments demonstrate that proposed methods notably enhance domain-specific ASR accuracy on real-life datasets.
• Score: 1.3791394805787949
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: End-to-end automatic speech recognition (E2E ASR) systems have significantly improved speech recognition through training on extensive datasets. Despite these advancements, they still struggle to accurately recognize domain specific words, such as proper nouns and technical terminologies. To address this problem, we propose a method to utilize the state-of-the-art Whisper without modifying its architecture, preserving its generalization performance while enabling it to leverage descriptions effectively. Moreover, we propose two additional training techniques to improve the domain specific ASR: decoder fine-tuning, and context perturbation. We also propose a method to use a Large Language Model (LLM) to generate descriptions with simple metadata, when descriptions are unavailable. Our experiments demonstrate that proposed methods notably enhance domain-specific ASR accuracy on real-life datasets, with LLM-generated descriptions outperforming human-crafted ones in effectiveness.

Related papers

• Retrieval-Augmented Speech Recognition Approach for Domain Challenges [24.337617843696286]
  Speech recognition systems often face challenges due to domain mismatch. Inspired by Retrieval-Augmented Generation (RAG) techniques for large language models (LLMs), this paper introduces a LLM-based retrieval-augmented speech recognition method.
  arXiv  Detail & Related papers  (2025-02-21T07:47:50Z)
• Developing Instruction-Following Speech Language Model Without Speech Instruction-Tuning Data [84.01401439030265]
  Recent end-to-end speech language models (SLMs) have expanded upon the capabilities of large language models (LLMs) We present a simple yet effective automatic process for creating speech-text pair data. Our model demonstrates general capabilities for speech-related tasks without the need for speech instruction-tuning data.
  arXiv  Detail & Related papers  (2024-09-30T07:01:21Z)
• LA-RAG:Enhancing LLM-based ASR Accuracy with Retrieval-Augmented Generation [15.520180125182756]
  Recent advancements in integrating speech information into large language models (LLMs) have significantly improved automatic speech recognition (ASR) accuracy. Existing methods often constrained by the capabilities of the speech encoders under varied acoustic conditions, such as accents. We propose LA-RAG, a novel Retrieval-Augmented Generation (RAG) paradigm for LLM-based ASR.
  arXiv  Detail & Related papers  (2024-09-13T07:28:47Z)
• LLM-DA: Data Augmentation via Large Language Models for Few-Shot Named Entity Recognition [67.96794382040547]
  $LLM-DA$ is a novel data augmentation technique based on large language models (LLMs) for the few-shot NER task. Our approach involves employing 14 contextual rewriting strategies, designing entity replacements of the same type, and incorporating noise injection to enhance robustness.
  arXiv  Detail & Related papers  (2024-02-22T14:19:56Z)
• Generative Context-aware Fine-tuning of Self-supervised Speech Models [54.389711404209415]
  We study the use of generative large language models (LLM) generated context information. We propose an approach to distill the generated information during fine-tuning of self-supervised speech models. We evaluate the proposed approach using the SLUE and Libri-light benchmarks for several downstream tasks: automatic speech recognition, named entity recognition, and sentiment analysis.
  arXiv  Detail & Related papers  (2023-12-15T15:46:02Z)
• ROSE: A Recognition-Oriented Speech Enhancement Framework in Air Traffic Control Using Multi-Objective Learning [6.60571587618006]
  Radio speech echo is a specific phenomenon in the air traffic control (ATC) domain, which degrades speech quality and impacts automatic speech recognition (ASR) accuracy. In this work, a time-domain recognition-oriented speech enhancement framework is proposed to improve speech intelligibility and advance ASR accuracy. The framework serves as a plug-and-play tool in ATC scenarios and does not require additional retraining of the ASR model.
  arXiv  Detail & Related papers  (2023-12-11T04:51:41Z)
• Noise-Robust Fine-Tuning of Pretrained Language Models via External Guidance [61.809732058101304]
  We introduce an innovative approach for fine-tuning PLMs using noisy labels. This approach incorporates the guidance of Large Language Models (LLMs) like ChatGPT. This guidance assists in accurately distinguishing between clean and noisy samples.
  arXiv  Detail & Related papers  (2023-11-02T09:20:38Z)
• SememeASR: Boosting Performance of End-to-End Speech Recognition against Domain and Long-Tailed Data Shift with Sememe Semantic Knowledge [58.979490858061745]
  We introduce sememe-based semantic knowledge information to speech recognition. Our experiments show that sememe information can improve the effectiveness of speech recognition. In addition, our further experiments show that sememe knowledge can improve the model's recognition of long-tailed data.
  arXiv  Detail & Related papers  (2023-09-04T08:35:05Z)
• A Deep Learning System for Domain-specific Speech Recognition [0.0]
  The author works with pre-trained DeepSpeech2 and Wav2Vec2 acoustic models to develop benefit-specific ASR systems. The best performance comes from a fine-tuned Wav2Vec2-Large-LV60 acoustic model with an external KenLM. The viability of using error prone ASR transcriptions as part of spoken language understanding (SLU) is also investigated.
  arXiv  Detail & Related papers  (2023-03-18T22:19:09Z)
• Visualizing Automatic Speech Recognition -- Means for a Better Understanding? [0.1868368163807795]
  We show how attribution methods, that we import from image recognition and suitably adapt to handle audio data, can help to clarify the working of ASR. Taking Speech Deep, an end-to-end model for ASR, as a case study, we show how these techniques help to visualize which features of the input are the most influential in determining the output.
  arXiv  Detail & Related papers  (2022-02-01T13:35:08Z)
• Seed Words Based Data Selection for Language Model Adaptation [11.59717828860318]
  We present an approach for automatically selecting sentences, from a text corpus, that match, both semantically and morphologically, a glossary of terms furnished by the user. The vocabulary of the baseline model is expanded and tailored, reducing the resulting OOV rate. Results using different metrics (OOV rate, WER, precision and recall) show the effectiveness of the proposed techniques.
  arXiv  Detail & Related papers  (2021-07-20T12:08:27Z)
• Improving Readability for Automatic Speech Recognition Transcription [50.86019112545596]
  We propose a novel NLP task called ASR post-processing for readability (APR) APR aims to transform the noisy ASR output into a readable text for humans and downstream tasks while maintaining the semantic meaning of the speaker. We compare fine-tuned models based on several open-sourced and adapted pre-trained models with the traditional pipeline method.
  arXiv  Detail & Related papers  (2020-04-09T09:26:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.