# Fast and Accurate Recurrent Neural Network Acoustic Models for Speech Recognition

arXiv: https://arxiv.org/abs/1507.06947 · HTML: https://mlsanity.com/paper/1507.06947/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Jul 2015.

## Built on by (most cited first, 26 of 26)

- [BigSSL: Exploring the Frontier of Large-Scale Semi-Supervised Learning for Automatic Speech Recognition](https://mlsanity.com/paper/2109.13226/) (2021)
- [CTC-based Compression for Direct Speech Translation](https://mlsanity.com/paper/2102.01578/) (2021)
- [Self-Ensemble Adversarial Training for Improved Robustness](https://mlsanity.com/paper/2203.09678/) (2022)
- [Why does CTC result in peaky behavior?](https://mlsanity.com/paper/2105.14849/) (2021)
- [Effect and Analysis of Large-scale Language Model Rescoring on Competitive ASR Systems](https://mlsanity.com/paper/2204.00212/) (2022)
- [Accelerating RNN-T Training and Inference Using CTC guidance](https://mlsanity.com/paper/2210.16481/) (2022)
- [Speechformer: Reducing Information Loss in Direct Speech Translation](https://mlsanity.com/paper/2109.04574/) (2021)
- [Reservoir Computing via Quantum Recurrent Neural Networks](https://mlsanity.com/paper/2211.02612/) (2022)
- [Generalist: Decoupling Natural and Robust Generalization](https://mlsanity.com/paper/2303.13813/) (2023)
- [YOLOv1 to YOLOv10: A comprehensive review of YOLO variants and their application in the agricultural domain](https://mlsanity.com/paper/2406.10139/) (2024)
- [Alignment Knowledge Distillation for Online Streaming Attention-based Speech Recognition](https://mlsanity.com/paper/2103.00422/) (2021)
- [Improving Adversarial Robustness via Mutual Information Estimation](https://mlsanity.com/paper/2207.12203/) (2022)
- [A Streaming End-to-End Framework For Spoken Language Understanding](https://mlsanity.com/paper/2105.10042/) (2021)
- [Delay-penalized transducer for low-latency streaming ASR](https://mlsanity.com/paper/2211.00490/) (2022)
- [TrimTail: Low-Latency Streaming ASR with Simple but Effective Spectrogram-Level Length Penalty](https://mlsanity.com/paper/2211.00522/) (2022)
- [Exploiting Large-scale Teacher-Student Training for On-device Acoustic Models](https://mlsanity.com/paper/2106.06126/) (2021)
- [Flexi-Transducer: Optimizing Latency, Accuracy and Compute forMulti-Domain On-Device Scenarios](https://mlsanity.com/paper/2104.02232/) (2021)
- [Optimizing Latency for Online Video CaptioningUsing Audio-Visual Transformers](https://mlsanity.com/paper/2108.02147/) (2021)
- [Biomembrane-based Memcapacitive Reservoir Computing System for Energy Efficient Temporal Data Processing](https://mlsanity.com/paper/2305.12025/) (2023)
- [Towards Better Meta-Initialization with Task Augmentation for Kindergarten-aged Speech Recognition](https://mlsanity.com/paper/2202.12326/) (2022)
- [Attacks as Defenses: Designing Robust Audio CAPTCHAs Using Attacks on Automatic Speech Recognition Systems](https://mlsanity.com/paper/2203.05408/) (2022)
- [Machine Learning based prediction of Glucose Levels in Type 1 Diabetes Patients with the use of Continuous Glucose Monitoring Data](https://mlsanity.com/paper/2302.12856/) (2023)
- [Parameter Interpolation Adversarial Training for Robust Image Classification](https://mlsanity.com/paper/2511.00836/) (2025)
- [Oracle Teacher: Leveraging Target Information for Better Knowledge Distillation of CTC Models](https://mlsanity.com/paper/2111.03664/) (2021)
- [Intelligent Multi-channel Meta-imagers for Accelerating Machine Vision](https://mlsanity.com/paper/2306.07365/) (2023)
- [ATWM: Defense against adversarial malware based on adversarial training](https://mlsanity.com/paper/2307.05095/) (2023)

## Abstract (arXiv)

We have recently shown that deep Long Short-Term Memory (LSTM) recurrent neural networks (RNNs) outperform feed forward deep neural networks (DNNs) as acoustic models for speech recognition. More recently, we have shown that the performance of sequence trained context dependent (CD) hidden Markov model (HMM) acoustic models using such LSTM RNNs can be equaled by sequence trained phone models initialized with connectionist temporal classification (CTC). In this paper, we present techniques that further improve performance of LSTM RNN acoustic models for large vocabulary speech recognition. We show that frame stacking and reduced frame rate lead to more accurate models and faster decoding. CD phone modeling leads to further improvements. We also present initial results for LSTM RNN models outputting words directly.