Quantization for OpenAI's Whisper Models: A Comparative Analysis

• URL: http://arxiv.org/abs/2503.09905v1
• Date: Wed, 12 Mar 2025 23:50:35 GMT
• Title: Quantization for OpenAI's Whisper Models: A Comparative Analysis
• Authors: Allison Andreyev,
• Abstract summary: This paper studies Whisper and two model variants: one optimized for live speech streaming and another for offline transcription.<n>Larger model variants exhibit increased latency and pose challenges for deployment on resource-constrained devices.<n>Results show that quantization reduces latency 19% and model size by 45%, while preserving transcription accuracy.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Automated speech recognition (ASR) models have gained prominence for applications such as captioning, speech translation, and live transcription. This paper studies Whisper and two model variants: one optimized for live speech streaming and another for offline transcription. Notably, these models have been found to generate hallucinated content, reducing transcription reliability. Furthermore, larger model variants exhibit increased latency and pose challenges for deployment on resource-constrained devices. This study analyzes the similarities and differences between three Whisper models, qualitatively examining their distinct capabilities. Next, this study quantifies the impact of model quantization on latency and evaluates its viability for edge deployment. Using the open source LibriSpeech dataset, this paper evaluates the word error rate (WER) along with latency analysis of whispercpp using 3 quantization methods (INT4, INT5, INT8). Results show that quantization reduces latency by 19\% and model size by 45\%, while preserving transcription accuracy. These findings provide insights into the optimal use cases of different Whisper models and edge device deployment possibilities. All code, datasets, and implementation details are available in a public GitHub repository: https://github.com/allisonandreyev/WhisperQuantization.git

Related papers

• Whisper Finetuning on Nepali Language [0.0]
  This research focuses on making an exhaustive and generalized dataset followed by fine-tuning OpenAI's Whisper models to improve transcription accuracy for the Nepali language. We leverage publicly available ASR datasets and self-recorded custom datasets with a diverse range of accents, dialects, and speaking styles further enriched through augmentation. Our approach outperforms Whisper's baseline models trained on Fleur's dataset, achieving WER reductions of up to 36.2% on the small and 23.8% on medium models.
  arXiv  Detail & Related papers  (2024-11-19T15:55:56Z)
• Contextual Biasing to Improve Domain-specific Custom Vocabulary Audio Transcription without Explicit Fine-Tuning of Whisper Model [0.0]
  OpenAI's Whisper Automated Speech Recognition model excels in generalizing across diverse datasets and domains. We propose a method to enhance transcription accuracy without explicit fine-tuning or altering model parameters.
  arXiv  Detail & Related papers  (2024-10-24T01:58:11Z)
• Distil-Whisper: Robust Knowledge Distillation via Large-Scale Pseudo Labelling [75.74809713084282]
  Distil-Whisper is designed to be paired with Whisper for speculative decoding, yielding a 2 times speed-up. Distil-Whisper is 5.8 times faster with 51% fewer parameters, while performing to within 1% WER on out-of-distribution test data. To facilitate further research in this domain, we make our training code, inference code and models publicly accessible.
  arXiv  Detail & Related papers  (2023-11-01T10:45:07Z)
• HyPoradise: An Open Baseline for Generative Speech Recognition with Large Language Models [81.56455625624041]
  We introduce the first open-source benchmark to utilize external large language models (LLMs) for ASR error correction. The proposed benchmark contains a novel dataset, HyPoradise (HP), encompassing more than 334,000 pairs of N-best hypotheses. LLMs with reasonable prompt and its generative capability can even correct those tokens that are missing in N-best list.
  arXiv  Detail & Related papers  (2023-09-27T14:44:10Z)
• Continual Learning for On-Device Speech Recognition using Disentangled Conformers [54.32320258055716]
  We introduce a continual learning benchmark for speaker-specific domain adaptation derived from LibriVox audiobooks. We propose a novel compute-efficient continual learning algorithm called DisentangledCL. Our experiments show that the DisConformer models significantly outperform baselines on general ASR.
  arXiv  Detail & Related papers  (2022-12-02T18:58:51Z)
• Analyzing Robustness of End-to-End Neural Models for Automatic Speech Recognition [11.489161072526677]
  We investigate robustness properties of pre-trained neural models for automatic speech recognition. In this work, we perform a robustness analysis of the pre-trained neural models wav2vec2, HuBERT and DistilHuBERT on the LibriSpeech and TIMIT datasets.
  arXiv  Detail & Related papers  (2022-08-17T20:00:54Z)
• Prediction of speech intelligibility with DNN-based performance measures [9.883633991083789]
  This paper presents a speech intelligibility model based on automatic speech recognition (ASR) It combines phoneme probabilities from deep neural networks (DNN) and a performance measure that estimates the word error rate from these probabilities. The proposed model performs almost as well as the label-based model and produces more accurate predictions than the baseline models.
  arXiv  Detail & Related papers  (2022-03-17T08:05:38Z)
• AES Systems Are Both Overstable And Oversensitive: Explaining Why And Proposing Defenses [66.49753193098356]
  We investigate the reason behind the surprising adversarial brittleness of scoring models. Our results indicate that autoscoring models, despite getting trained as "end-to-end" models, behave like bag-of-words models. We propose detection-based protection models that can detect oversensitivity and overstability causing samples with high accuracies.
  arXiv  Detail & Related papers  (2021-09-24T03:49:38Z)
• Bridging the Gap Between Clean Data Training and Real-World Inference for Spoken Language Understanding [76.89426311082927]
  Existing models are trained on clean data, which causes a textitgap between clean data training and real-world inference. We propose a method from the perspective of domain adaptation, by which both high- and low-quality samples are embedding into similar vector space. Experiments on the widely-used dataset, Snips, and large scale in-house dataset (10 million training examples) demonstrate that this method not only outperforms the baseline models on real-world (noisy) corpus but also enhances the robustness, that is, it produces high-quality results under a noisy environment.
  arXiv  Detail & Related papers  (2021-04-13T17:54:33Z)
• Switching Variational Auto-Encoders for Noise-Agnostic Audio-visual Speech Enhancement [26.596930749375474]
  We introduce the use of a latent sequential variable with Markovian dependencies to switch between different VAE architectures through time. We derive the corresponding variational expectation-maximization algorithm to estimate the parameters of the model and enhance the speech signal.
  arXiv  Detail & Related papers  (2021-02-08T11:45:02Z)
• Parameter Space Factorization for Zero-Shot Learning across Tasks and Languages [112.65994041398481]
  We propose a Bayesian generative model for the space of neural parameters. We infer the posteriors over such latent variables based on data from seen task-language combinations. Our model yields comparable or better results than state-of-the-art, zero-shot cross-lingual transfer methods.
  arXiv  Detail & Related papers  (2020-01-30T16:58:56Z)

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.