Saved in:
Bibliographic Details
Main Authors: Timmel, Vincenzo, Paonessa, Claudio, Kakooee, Reza, Vogel, Manfred, Perruchoud, Daniel
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.15726
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912339860652032
author Timmel, Vincenzo
Paonessa, Claudio
Kakooee, Reza
Vogel, Manfred
Perruchoud, Daniel
author_facet Timmel, Vincenzo
Paonessa, Claudio
Kakooee, Reza
Vogel, Manfred
Perruchoud, Daniel
contents This paper presents a new approach to fine-tuning OpenAI's Whisper model for low-resource languages by introducing a novel data generation method that converts sentence-level data into a long-form corpus, using Swiss German as a case study. Non-sentence-level data, which could improve the performance of long-form audio, is difficult to obtain and often restricted by copyright laws. Our method bridges this gap by transforming more accessible sentence-level data into a format that preserves the model's ability to handle long-form audio and perform segmentation without requiring non-sentence-level data. Our data generation process improves performance in several real-world applications and leads to the development of a new state-of-the-art speech-to-text (STT) model for Swiss German. We compare our model with a non-fine-tuned Whisper and our previous state-of-the-art Swiss German STT models, where our new model achieves higher BLEU scores. Our results also indicate that the proposed method is adaptable to other low-resource languages, supported by written guidance and code that allows the creation of fine-tuned Whisper models, which keep segmentation capabilities and allow the transcription of longer audio files using only sentence-level data with high quality.
format Preprint
id arxiv_https___arxiv_org_abs_2412_15726
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Fine-tuning Whisper on Low-Resource Languages for Real-World Applications
Timmel, Vincenzo
Paonessa, Claudio
Kakooee, Reza
Vogel, Manfred
Perruchoud, Daniel
Computation and Language
Audio and Speech Processing
This paper presents a new approach to fine-tuning OpenAI's Whisper model for low-resource languages by introducing a novel data generation method that converts sentence-level data into a long-form corpus, using Swiss German as a case study. Non-sentence-level data, which could improve the performance of long-form audio, is difficult to obtain and often restricted by copyright laws. Our method bridges this gap by transforming more accessible sentence-level data into a format that preserves the model's ability to handle long-form audio and perform segmentation without requiring non-sentence-level data. Our data generation process improves performance in several real-world applications and leads to the development of a new state-of-the-art speech-to-text (STT) model for Swiss German. We compare our model with a non-fine-tuned Whisper and our previous state-of-the-art Swiss German STT models, where our new model achieves higher BLEU scores. Our results also indicate that the proposed method is adaptable to other low-resource languages, supported by written guidance and code that allows the creation of fine-tuned Whisper models, which keep segmentation capabilities and allow the transcription of longer audio files using only sentence-level data with high quality.
title Fine-tuning Whisper on Low-Resource Languages for Real-World Applications
topic Computation and Language
Audio and Speech Processing
url https://arxiv.org/abs/2412.15726