# Improving RNN Transducer Based ASR with Auxiliary Tasks Source: [Improving RNN Transducer Based ASR with Auxiliary Tasks]() Collection index: [llms.txt]() Data updated: 2026-01-01T13:06:46.366Z Authors: Chunxi Liu, Frank Zhang, Duc Le, Suyoun Kim, Yatharth Saraf, Geoffrey Zweig Published: 2020-11-05T21:46:32.000Z ## Abstract End-to-end automatic speech recognition (ASR) models with a single neural network have recently demonstrated state-of-the-art results compared to conventional hybrid speech recognizers. Specifically, recurrent neural network transducer (RNN-T) has shown competitive ASR performance on various benchmarks. In this work, we examine ways in which RNN-T can achieve better ASR accuracy via performing auxiliary tasks. We propose (i) using the same auxiliary task as primary RNN-T ASR task, and (ii) performing context-dependent graphemic state prediction as in conventional hybrid modeling. In transcribing social media videos with varying training data size, we first evaluate the streaming ASR performance on three languages: Romanian, Turkish and German. We find that both proposed methods provide consistent improvements. Next, we observe that both auxiliary tasks demonstrate efficacy in learning deep transformer encoders for RNN-T criterion, thus achieving competitive results - 2.0%/4.2% WER on LibriSpeech test-clean/other - as compared to prior top performing models. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Speech Recognition on Turkish video dataset (clean)](): Speech Recognition; Turkish video dataset (clean); WER: 16.1. - [Speech Recognition on Turkish video dataset (noisy)](): Speech Recognition; Turkish video dataset (noisy); WER: 18.1. - [Speech Recognition on German video dataset (clean)](): Speech Recognition; German video dataset (clean); WER: 11.2. - [Speech Recognition on German video dataset (noisy)](): Speech Recognition; German video dataset (noisy); WER: 12.3. - [Automatic Speech Recognition on LibriSpeech clean (test)](): Automatic Speech Recognition; LibriSpeech clean (test); WER: 2. - [Automatic Speech Recognition on LibriSpeech (test-other)](): Automatic Speech Recognition; LibriSpeech (test-other); WER: 4.2. ## Original sources and implementation [Original paper]() [PDF]()