# Paraformer: Fast and Accurate Parallel Transformer for Non-autoregressive End-to-End Speech Recognition

arXiv: https://arxiv.org/abs/2206.08317 · HTML: https://mlsanity.com/paper/2206.08317/

## Key facts

- Cited by 144 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 32 references that are in the map.
- First posted on arXiv 16 Jun 2022.

## Builds on (most cited first, 27 of 27)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [Speech Recognition with Deep Recurrent Neural Networks](https://mlsanity.com/paper/1303.5778/) (2013)
- [ESPnet: End-to-End Speech Processing Toolkit](https://mlsanity.com/paper/1804.00015/) (2018)
- [Listen, Attend and Spell](https://mlsanity.com/paper/1508.01211/) (2015)
- [Non-Autoregressive Neural Machine Translation](https://mlsanity.com/paper/1711.02281/) (2017)
- [AISHELL-1: An Open-Source Mandarin Speech Corpus and A Speech Recognition Baseline](https://mlsanity.com/paper/1709.05522/) (2017)
- [OpenNMT: Open-source Toolkit for Neural Machine Translation](https://mlsanity.com/paper/1709.03815/) (2017)
- [Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement](https://mlsanity.com/paper/1802.06901/) (2018)
- [AISHELL-2: Transforming Mandarin ASR Research Into Industrial Scale](https://mlsanity.com/paper/1808.10583/) (2018)
- [Glancing Transformer for Non-Autoregressive Neural Machine Translation](https://mlsanity.com/paper/2008.07905/) (2020)
- [CIF: Continuous Integrate-and-Fire for End-to-End Speech Recognition](https://mlsanity.com/paper/1905.11235/) (2019)
- [Mask CTC: Non-Autoregressive End-to-End ASR with CTC and Mask Predict](https://mlsanity.com/paper/2005.08700/) (2020)
- [Minimum Word Error Rate Training for Attention-based Sequence-to-Sequence Models](https://mlsanity.com/paper/1712.01818/) (2017)
- [Improved Mask-CTC for Non-Autoregressive End-to-End ASR](https://mlsanity.com/paper/2010.13270/) (2020)
- [Fast End-to-End Speech Recognition via Non-Autoregressive Models and Cross-Modal Knowledge Transferring from BERT](https://mlsanity.com/paper/2102.07594/) (2021)
- [Spike-Triggered Non-Autoregressive Transformer for End-to-End Speech Recognition](https://mlsanity.com/paper/2005.07903/) (2020)
- [CASS-NAT: CTC Alignment-based Single Step Non-autoregressive Transformer for Speech Recognition](https://mlsanity.com/paper/2010.14725/) (2020)
- [Non-Autoregressive Transformer ASR with CTC-Enhanced Decoder Input](https://mlsanity.com/paper/2010.15025/) (2020)
- [Improving non-autoregressive end-to-end speech recognition with pre-trained acoustic and language models](https://mlsanity.com/paper/2201.10103/) (2022)
- [SAN-M: Memory Equipped Self-Attention for End-to-End Speech Recognition](https://mlsanity.com/paper/2006.01713/) (2020)
- [Improving CTC-based speech recognition via knowledge transferring from pre-trained language models](https://mlsanity.com/paper/2203.03582/) (2022)
- [Streaming Chunk-Aware Multihead Attention for Online End-to-End Speech Recognition](https://mlsanity.com/paper/2006.01712/) (2020)
- [Streaming End-to-End ASR based on Blockwise Non-Autoregressive Models](https://mlsanity.com/paper/2107.09428/) (2021)
- [Universal ASR: Unifying Streaming and Non-Streaming ASR Using a Single Encoder-Decoder Model](https://mlsanity.com/paper/2010.14099/) (2020)
- [An Improved Single Step Non-autoregressive Transformer for Automatic Speech Recognition](https://mlsanity.com/paper/2106.09885/) (2021)
- [Non-autoregressive Transformer with Unified Bidirectional Decoder for Automatic Speech Recognition](https://mlsanity.com/paper/2109.06684/) (2021)

## Built on by (most cited first, 30 of 144)

- [CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models](https://mlsanity.com/paper/2412.10117/) (2024)
- [Kimi-Audio Technical Report](https://mlsanity.com/paper/2504.18425/) (2025)
- [Freeze-Omni: A Smart and Low Latency Speech-to-speech Dialogue Model with Frozen LLM](https://mlsanity.com/paper/2411.00774/) (2024)
- [FunAudioLLM: Voice Understanding and Generation Foundation Models for Natural Interaction Between Humans and LLMs](https://mlsanity.com/paper/2407.04051/) (2024)
- [FunASR: A Fundamental End-to-End Speech Recognition Toolkit](https://mlsanity.com/paper/2305.11013/) (2023)
- [Unified Multimodal Understanding and Generation Models: Advances, Challenges, and Opportunities](https://mlsanity.com/paper/2505.02567/) (2025)
- [Step-Audio: Unified Understanding and Generation in Intelligent Speech Interaction](https://mlsanity.com/paper/2502.11946/) (2025)
- [LauraGPT: Listen, Attend, Understand, and Regenerate Audio with GPT](https://mlsanity.com/paper/2310.04673/) (2023)
- [MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction](https://mlsanity.com/paper/2604.27393/) (2026)
- [Llasa: Scaling Train-Time and Inference-Time Compute for Llama-based Speech Synthesis](https://mlsanity.com/paper/2502.04128/) (2025)
- [MiMo-Audio: Audio Language Models are Few-Shot Learners](https://mlsanity.com/paper/2512.23808/) (2025)
- [Enhance Reasoning for Large Language Models in the Game Werewolf](https://mlsanity.com/paper/2402.02330/) (2024)
- [Baichuan-Omni Technical Report](https://mlsanity.com/paper/2410.08565/) (2024)
- [Transformers in Speech Processing: A Survey](https://mlsanity.com/paper/2303.11607/) (2023)
- [Recurrent Drafter for Fast Speculative Decoding in Large Language Models](https://mlsanity.com/paper/2403.09919/) (2024)
- [VITA-Audio: Fast Interleaved Cross-Modal Token Generation for Efficient Large Speech-Language Model](https://mlsanity.com/paper/2505.03739/) (2025)
- [ZipVoice: Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching](https://mlsanity.com/paper/2506.13053/) (2025)
- [VibeVoice Technical Report](https://mlsanity.com/paper/2508.19205/) (2025)
- [IndexTTS: An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System](https://mlsanity.com/paper/2502.05512/) (2025)
- [OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Models](https://mlsanity.com/paper/2604.00688/) (2026)
- [M2-omni: Advancing Omni-MLLM for Comprehensive Modality Support with Competitive Performance](https://mlsanity.com/paper/2502.18778/) (2025)
- [SeACo-Paraformer: A Non-Autoregressive ASR System with Flexible and Effective Hotword Customization Ability](https://mlsanity.com/paper/2308.03266/) (2023)
- [MoonCast: High-Quality Zero-Shot Podcast Generation](https://mlsanity.com/paper/2503.14345/) (2025)
- [X-Streamer: Unified Human World Modeling with Audiovisual Interaction](https://mlsanity.com/paper/2509.21574/) (2025)
- [Comparing Discrete and Continuous Space LLMs for Speech Recognition](https://mlsanity.com/paper/2409.00800/) (2024)
- [Differentiable Reward Optimization for LLM based TTS system](https://mlsanity.com/paper/2507.05911/) (2025)
- [NVSpeech: An Integrated and Scalable Pipeline for Human-Like Speech Modeling with Paralinguistic Vocalizations](https://mlsanity.com/paper/2508.04195/) (2025)
- [SpeechJudge: Towards Human-Level Judgment for Speech Naturalness](https://mlsanity.com/paper/2511.07931/) (2025)
- [Unveiling the Potential of LLM-Based ASR on Chinese Open-Source Datasets](https://mlsanity.com/paper/2405.02132/) (2024)
- [Easy Turn: Integrating Acoustic and Linguistic Modalities for Robust Turn-Taking in Full-Duplex Spoken Dialogue Systems](https://mlsanity.com/paper/2509.23938/) (2025)

## Abstract (arXiv)

Transformers have recently dominated the ASR field. Although able to yield good performance, they involve an autoregressive (AR) decoder to generate tokens one by one, which is computationally inefficient. To speed up inference, non-autoregressive (NAR) methods, e.g. single-step NAR, were designed, to enable parallel generation. However, due to an independence assumption within the output tokens, performance of single-step NAR is inferior to that of AR models, especially with a large-scale corpus. There are two challenges to improving single-step NAR: Firstly to accurately predict the number of output tokens and extract hidden variables; secondly, to enhance modeling of interdependence between output tokens. To tackle both challenges, we propose a fast and accurate parallel transformer, termed Paraformer. This utilizes a continuous integrate-and-fire based predictor to predict the number of tokens and generate hidden variables. A glancing language model (GLM) sampler then generates semantic embeddings to enhance the NAR decoder's ability to model context interdependence. Finally, we design a strategy to generate negative samples for minimum word error rate training to further improve performance. Experiments using the public AISHELL-1, AISHELL-2 benchmark, and an industrial-level 20,000 hour task demonstrate that the proposed Paraformer can attain comparable performance to the state-of-the-art AR transformer, with more than 10x speedup.