Two-pass Decoding And Cross-adaptation Based System Combination Of End-to-end Conformer And Hybrid TDNN ASR Systems

Abstract

Fundamental modelling differences between hybrid and end-to-end (E2E) automatic speech recognition (ASR) systems create large diversity and complementarity among them. This paper investigates multi-pass rescoring and cross adaptation based system combination approaches for hybrid TDNN and Conformer E2E ASR systems. In multi-pass rescoring, state-of-the-art hybrid LF-MMI trained CNN-TDNN system featuring speed perturbation, SpecAugment and Bayesian learning hidden unit contributions (LHUC) speaker adaptation was used to produce initial N-best outputs before being rescored by the speaker adapted Conformer system using a 2-way cross system score interpolation. In cross adaptation, the hybrid CNN-TDNN system was adapted to the 1-best output of the Conformer system or vice versa. Experiments on the 300-hour Switchboard corpus suggest that the combined systems derived using either of the two system combination approaches outperformed the individual systems. The best combined system obtained

Stats

• citations6
• S2 citations—
• github stars0
• HF likes0
• heat score6.34
• arxiv keycui2022two

Related papers

• Have Best Of Both Worlds: Two-pass Hybrid And E2E Cascading Framework For Speech Recognition (2021)6.34
• Audio-attention Discriminative Language Model For ASR Rescoring (2019)9.23
• Conformer-based Hybrid ASR System For Switchboard Dataset (2021)9.41
• Multiple-hypothesis Ctc-based Semi-supervised Adaptation Of End-to-end Speech Recognition (2021)5.84
• Combining Frame-synchronous And Label-synchronous Systems For Speech Recognition (2021)0.00
• 4D ASR: Joint Modeling Of CTC, Attention, Transducer, And Mask-predict Decoders (2022)7.50
• Advancing CTC-CRF Based End-to-end Speech Recognition With Wordpieces And Conformers (2021)0.00
• Confidence Score Based Conformer Speaker Adaptation For Speech Recognition (2022)8.09