Typhoon ASR Real-time: FastConformer-Transducer for Thai Automatic Speech Recognition

TL;DR

This paper introduces Typhoon ASR Real-time, a low-latency FastConformer-Transducer model for Thai speech recognition, achieving high accuracy with significantly reduced computational cost and supporting dialect adaptation and standardized evaluation.

Contribution

The paper presents a compact FastConformer-Transducer model for Thai ASR, a normalization pipeline for consistent training, a dialect adaptation method, and a new benchmark dataset for reproducibility.

Findings

45x reduction in computational cost compared to Whisper Large-v3

Achieves comparable accuracy to large offline models

Introduces a standardized Thai ASR benchmark dataset

Abstract

Large encoder-decoder models like Whisper achieve strong offline transcription but remain impractical for streaming applications due to high latency. However, due to the accessibility of pre-trained checkpoints, the open Thai ASR landscape remains dominated by these offline architectures, leaving a critical gap in efficient streaming solutions. We present Typhoon ASR Real-time, a 115M-parameter FastConformer-Transducer model for low-latency Thai speech recognition. We demonstrate that rigorous text normalization can match the impact of model scaling: our compact model achieves a 45x reduction in computational cost compared to Whisper Large-v3 while delivering comparable accuracy. Our normalization pipeline resolves systemic ambiguities in Thai transcription --including context-dependent number verbalization and repetition markers (mai yamok) --creating consistent training targets. We further introduce a two-stage curriculum learning approach for Isan (north-eastern) dialect adaptation that preserves Central Thai performance. To address reproducibility challenges in Thai ASR, we release the Typhoon ASR Benchmark, a gold-standard human-labeled datasets with transcriptions following established Thai linguistic conventions, providing standardized evaluation protocols for the research community.