# Benchmarking Rotary Position Embeddings for Automatic Speech Recognition

arXiv: https://arxiv.org/abs/2501.06051 · HTML: https://mlsanity.com/paper/2501.06051/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 29 references that are in the map.
- First posted on arXiv 10 Jan 2025.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://mlsanity.com/paper/1901.02860/) (2019)
- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [ESPnet: End-to-End Speech Processing Toolkit](https://mlsanity.com/paper/1804.00015/) (2018)
- [VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation](https://mlsanity.com/paper/2101.00390/) (2021)
- [SpeechBrain: A General-Purpose Speech Toolkit](https://mlsanity.com/paper/2106.04624/) (2021)
- [Effective Long-Context Scaling of Foundation Models](https://mlsanity.com/paper/2309.16039/) (2023)
- [Joint CTC-Attention based End-to-End Speech Recognition using Multi-task Learning](https://mlsanity.com/paper/1609.06773/) (2016)
- [NeMo: a toolkit for building AI applications using Neural Modules](https://mlsanity.com/paper/1909.09577/) (2019)
- [Spirit LM: Interleaved Spoken and Written Language Model](https://mlsanity.com/paper/2402.05755/) (2024)
- [WeNet: Production oriented Streaming and Non-streaming End-to-End Speech Recognition Toolkit](https://mlsanity.com/paper/2102.01547/) (2021)
- [Libriheavy: a 50,000 hours ASR corpus with punctuation casing and context](https://mlsanity.com/paper/2309.08105/) (2023)
- [Open-Source Conversational AI with SpeechBrain 1.0](https://mlsanity.com/paper/2407.00463/) (2024)
- [On the Usefulness of Self-Attention for Automatic Speech Recognition with Transformers](https://mlsanity.com/paper/2011.04906/) (2020)
- [Moonshine: Speech Recognition for Live Transcription and Voice Commands](https://mlsanity.com/paper/2410.15608/) (2024)
- [SummaryMixing: A Linear-Complexity Alternative to Self-Attention for Speech Recognition and Understanding](https://mlsanity.com/paper/2307.07421/) (2023)
- [Conformer-based End-to-end Speech Recognition With Rotary Position Embedding](https://mlsanity.com/paper/2107.05907/) (2021)

## Built on by (most cited first, 1 of 1)

- [CAM~OES: A Comprehensive Automatic Speech Recognition Benchmark for European Portuguese](https://mlsanity.com/paper/2508.19721/) (2025)

## Abstract (arXiv)

Self-attention relies on positional embeddings to encode input order. Relative Position (RelPos) embeddings are widely used in Automatic Speech Recognition (ASR). However, RelPos has quadratic time complexity to input length and is often incompatible with fast GPU implementations of attention. In contrast, Rotary Positional Embedding (RoPE) rotates each input vector based on its absolute position, taking linear time to sequence length, implicitly encoding relative distances through self-attention dot products. Thus, it is usually compatible with efficient attention. However, its use in ASR remains underexplored. This work evaluates RoPE across diverse ASR tasks with training data ranging from 100 to 50,000 hours, covering various speech types (read, spontaneous, clean, noisy) and different accents in both streaming and non-streaming settings. ASR error rates are similar or better than RelPos, while training time is reduced by up to 21%. Code is available via the SpeechBrain toolkit.