Performance Improvement of Language-Queried Audio Source Separation Based on Caption Augmentation From Large Language Models for DCASE Challenge 2024 Task 9
- URL: http://arxiv.org/abs/2406.11248v1
- Date: Mon, 17 Jun 2024 06:19:14 GMT
- Title: Performance Improvement of Language-Queried Audio Source Separation Based on Caption Augmentation From Large Language Models for DCASE Challenge 2024 Task 9
- Authors: Do Hyun Lee, Yoonah Song, Hong Kook Kim,
- Abstract summary: We present a prompt-engineering-based text-augmentation approach applied to a language-queried audio source separation (LASS) task.
To enhance the performance of LASS, the proposed approach utilizes large language models (LLMs) to generate multiple captions corresponding to each sentence of the training dataset.
- Score: 4.328586290529485
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: We present a prompt-engineering-based text-augmentation approach applied to a language-queried audio source separation (LASS) task. To enhance the performance of LASS, the proposed approach utilizes large language models (LLMs) to generate multiple captions corresponding to each sentence of the training dataset. To this end, we first perform experiments to identify the most effective prompts for caption augmentation with a smaller number of captions. A LASS model trained with these augmented captions demonstrates improved performance on the DCASE 2024 Task 9 validation set compared to that trained without augmentation. This study highlights the effectiveness of LLM-based caption augmentation in advancing language-queried audio source separation.
Related papers
- AudioSetMix: Enhancing Audio-Language Datasets with LLM-Assisted Augmentations [1.2101820447447276]
Multi-modal learning in the audio-language domain has seen significant advancements in recent years.
However, audio-language learning faces challenges due to limited and lower-quality data compared to image-language tasks.
Our method systematically generates audio-caption pairs by augmenting audio clips with natural language labels and corresponding audio signal processing operations.
This scalable method produces AudioSetMix, a high-quality training dataset for text-and-audio related models.
arXiv Detail & Related papers (2024-05-17T21:08:58Z) - A Tale of Two Languages: Large-Vocabulary Continuous Sign Language Recognition from Spoken Language Supervision [74.972172804514]
We introduce a multi-task Transformer model, CSLR2, that is able to ingest a signing sequence and output in a joint embedding space between signed language and spoken language text.
New dataset annotations provide continuous sign-level annotations for six hours of test videos, and will be made publicly available.
Our model significantly outperforms the previous state of the art on both tasks.
arXiv Detail & Related papers (2024-05-16T17:19:06Z) - Generative linguistic representation for spoken language identification [17.9575874225144]
We explore the utilization of the decoder-based network from the Whisper model to extract linguistic features.
We devised two strategies - one based on the language embedding method and the other focusing on direct optimization of LID outputs.
We conducted experiments on the large-scale multilingual datasets MLS, VoxLingua107, and CommonVoice to test our approach.
arXiv Detail & Related papers (2023-12-18T06:40:24Z) - HowToCaption: Prompting LLMs to Transform Video Annotations at Scale [77.02631712558251]
We propose to leverage the capability of large language models (LLMs) to obtain fine-grained video descriptions aligned with videos.
We apply our method to the subtitles of the HowTo100M dataset, creating a new large-scale dataset, HowToCaption.
Our evaluation shows that the resulting captions not only significantly improve the performance over many different benchmark datasets for text-video retrieval.
arXiv Detail & Related papers (2023-10-07T19:32:55Z) - End-to-End Speech Recognition Contextualization with Large Language
Models [25.198480789044346]
We introduce a novel method for contextualizing speech recognition models incorporating Large Language Models (LLMs)
We provide audio features, along with optional text tokens for context, to train the system to complete transcriptions in a decoder-only fashion.
Our empirical results demonstrate a significant improvement in performance, with a 6% WER reduction when additional textual context is provided.
arXiv Detail & Related papers (2023-09-19T20:28:57Z) - Improving Natural-Language-based Audio Retrieval with Transfer Learning
and Audio & Text Augmentations [7.817685358710508]
We propose a system to project recordings and textual descriptions into a shared audio-caption space.
Our results show that the used augmentations strategies reduce overfitting and improve retrieval performance.
We further show that pre-training the system on the AudioCaps dataset leads to additional improvements.
arXiv Detail & Related papers (2022-08-24T11:54:42Z) - Transducer-based language embedding for spoken language identification [38.60303603000269]
The acoustic and linguistic features are important cues for the spoken language identification task.
Recent advanced LID systems mainly use acoustic features that lack the usage of explicit linguistic feature encoding.
We propose a novel transducer-based language embedding approach for LID tasks by integrating an RNN transducer model into a language embedding framework.
arXiv Detail & Related papers (2022-04-08T07:23:43Z) - BLIP: Bootstrapping Language-Image Pre-training for Unified
Vision-Language Understanding and Generation [86.4572981982407]
We propose BLIP, a new vision-language framework which transfers flexibly to both vision-language understanding and generation tasks.
BLIP effectively utilizes the noisy web data by bootstrapping the captions, where a captioner generates synthetic captions and a filter removes the noisy ones.
BLIP also demonstrates strong generalization ability when directly transferred to video-language tasks in a zero-shot manner.
arXiv Detail & Related papers (2022-01-28T12:49:48Z) - WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech
Processing [102.45426364965887]
We propose a new pre-trained model, WavLM, to solve full-stack downstream speech tasks.
WavLM is built based on the HuBERT framework, with an emphasis on both spoken content modeling and speaker identity preservation.
We scale up the training dataset from 60k hours to 94k hours of public audio data, and optimize its training procedure for better representation extraction.
arXiv Detail & Related papers (2021-10-26T17:55:19Z) - UniSpeech-SAT: Universal Speech Representation Learning with Speaker
Aware Pre-Training [72.004873454347]
Two methods are introduced for enhancing the unsupervised speaker information extraction.
Experiment results on SUPERB benchmark show that the proposed system achieves state-of-the-art performance.
We scale up training dataset to 94 thousand hours public audio data and achieve further performance improvement.
arXiv Detail & Related papers (2021-10-12T05:43:30Z) - SPLAT: Speech-Language Joint Pre-Training for Spoken Language
Understanding [61.02342238771685]
Spoken language understanding requires a model to analyze input acoustic signal to understand its linguistic content and make predictions.
Various pre-training methods have been proposed to learn rich representations from large-scale unannotated speech and text.
We propose a novel semi-supervised learning framework, SPLAT, to jointly pre-train the speech and language modules.
arXiv Detail & Related papers (2020-10-05T19:29:49Z)
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.