Updated Corpora and Benchmarks for Long-Form Speech Recognition

2023/09/26 by Jennifer Drexler Fox, Fox, Jennifer Drexler, Desh Raj +9 · 1 citation
Computer Science · #Audio and Speech Processing (eess.AS) #Computation and Language (cs.CL) #FOS: Computer and information sciences #FOS: Electrical engineering #Sound (cs.SD) #Speech Recognition and Synthesis #Speech and Audio Processing #Speech and dialogue systems #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2309.15013

openalex publication_date 2023/09/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The vast majority of ASR research uses corpora in which both the training and test data have been pre-segmented into utterances. In most real-word ASR use-cases, however, test audio is not segmented, leading to a mismatch between inference-time conditions and models trained on segmented utterances. In this paper, we re-release three standard ASR corpora - TED-LIUM 3, Gigapeech, and VoxPopuli-en - with updated transcription and alignments to enable their use for long-form ASR research. We use these reconstituted corpora to study the train-test mismatch problem for transducers and attention-based encoder-decoders (AEDs), confirming that AEDs are more susceptible to this issue. Finally, we benchmark a simple long-form training for these models, showing its efficacy for model robustness under this domain shift.

Cited by

• Segmental Attention Decoding With Long Form Acoustic Encodings

Related

• Investigating End-to-End ASR Architectures for Long Form Audio Transcription
• Improved Long-Form Speech Recognition by Jointly Modeling the Primary and Non-primary Speakers
• Better Pseudo-labeling with Multi-ASR Fusion and Error Correction by SpeechLLM
• Towards Pretraining Robust ASR Foundation Model with Acoustic-Aware Data Augmentation
• Adapting Large Language Model with Speech for Fully Formatted End-to-End Speech Recognition