OpenCodePapers

speech-recognition-on-ami

Speech Recognition
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeWord Error Rate (WER)ModelNameReleaseDate
NLE: Non-autoregressive LLM-based ASR by Transcript Editing✓ Link7.64ibm-granite/granite-speech-4.1-2b-nar2026-04-01
✓ Link7.72ibm-granite/granite-speech-4.1-2b2026-04-29
Introducing Cohere-transcribe: state-of-the-art speech recognition7.8CohereLabs/cohere-transcribe-03-20262026-03-26
Boson AI Launches Higgs STT 3 Speech-to-Text Model✓ Link7.86bosonai/higgs-audio-v3-stt2026-03-18
Azure Speech at Build 2026: Powering Voice Agents with Real-Time and Life-like Experiences7.93microsoft/azure-speech-06-20262026-06-04
CrisperWhisper: Accurate Timestamps on Verbatim Speech Transcriptions✓ Link8.04nyrahealth/CrisperWhisper2024-08-29
✓ Link8.12ibm-granite/granite-4.0-1b-speech2026-03-06
Granite-speech: open-source speech-aware LLMs with strong English ASR capabilities✓ Link8.43ibm-granite/granite-speech-3.3-8b2025-06-19
8.48OpenMOSS-Team/MOSS-Transcribe-preview-2B2026-06-26
Granite-speech: open-source speech-aware LLMs with strong English ASR capabilities✓ Link8.53ibm-granite/granite-speech-3.3-2b2025-06-19
Data-Efficient On-Policy Distillation for Automatic Speech Recognition✓ Link8.79AutoArk-AI/ARK-ASR-3B2026-06-22
Introducing Zoom AI Services8.9zoom/scribe_v12026-03-10
✓ Link9.05nvidia/canary-qwen-2.5b2025-07-17
Qwen3-ASR Technical Report✓ Link9.26Qwen/Qwen3-ASR-1.7B2026-01-28
Boson AI Launches Higgs STT 3 Speech-to-Text Model✓ Link9.35bosonai/higgs-audio-v3-8b-stt-v22026-04-27
Introducing Resonant-1 and Resonant-1-flash9.54reson8/resonant-1-flash2026-04-08
Introducing Resonant-1 and Resonant-1-flash9.63reson8/resonant-12026-04-08
Moonshine v2: Ergodic Streaming Encoder ASR for Latency-Critical Speech Applications✓ Link9.75usefulsensors/moonshine-streaming-medium2026-02-12
Data-Efficient On-Policy Distillation for Automatic Speech Recognition✓ Link9.81AutoArk-AI/ARK-ASR-0.6B2026-05-25
Avalon: ASR for Human–AI Interaction10.19aquavoice/avalon-v1-en2025-08-22
Efficient Sequence Transduction by Jointly Predicting Tokens and Durations✓ Link10.4nvidia/parakeet-tdt-0.6b-v22025-05-01
Pulse STT: Fast, Accurate Speech-to-Text10.42smallestai/pulse2026-01-28
Qwen3-ASR Technical Report✓ Link10.57Qwen/Qwen3-ASR-0.6B2026-01-28
Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST✓ Link10.58nvidia/parakeet-tdt-0.6b-v32025-08-14
Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs10.69microsoft/Phi-4-multimodal-instruct2025-02-24
Streaming Sequence-to-Sequence Learning with Delayed Streams Modeling✓ Link11.57kyutai/stt-2.6b-en2025-06-06
Moonshine v2: Ergodic Streaming Encoder ASR for Latency-Critical Speech Applications✓ Link11.69usefulsensors/moonshine-streaming-small2026-02-12
Training and Inference Efficiency of Encoder-Decoder Speech Models✓ Link12.11nvidia/canary-1b-flash2025-03-07
Less is More: Accurate Speech Recognition & Translation without Web-Scale Data✓ Link12.42nvidia/canary-1b2024-02-07
Introducing Universal-3 Pro: A new class of speech language model optimized for Voice AI12.56assemblyai/universal-3-pro2026-02-03
OWSM v4: Improving Open Whisper-Style Speech Models via Data Scaling and Cleaning✓ Link12.58espnet/owsm_ctc_v4_1B2025-01-16
Ursa 2: Elevating speech recognition across 50+ languages13.31speechmatics/enhanced2024-10-11
Robust Knowledge Distillation via Large-Scale Pseudo Labelling✓ Link13.36distil-whisper/distil-large-v3.52024-12-05
Training and Inference Efficiency of Encoder-Decoder Speech Models✓ Link13.66nvidia/canary-180m-flash2025-03-11
Zipformer: A faster and better encoder for automatic speech recognition✓ Link13.96soundsgoodai/Zipformer-transducer-XL-290M2026-05-12
Voxtral✓ Link14.32mistralai/Voxtral-Small-24B-25072025-07-15
Efficient Sequence Transduction by Jointly Predicting Tokens and Durations✓ Link14.7nvidia/parakeet-tdt_ctc-110m2024-09-17
Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST✓ Link14.71nvidia/canary-1b-v22025-08-14
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link14.76nvidia/parakeet-ctc-1.1b2023-12-28
Efficient Sequence Transduction by Jointly Predicting Tokens and Durations✓ Link14.81nvidia/parakeet-tdt-1.1b2024-01-25
LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation✓ Link14.81efficient-speech/lite-whisper-large-v3-acc2025-02-26
Robust Speech Recognition via Large-Scale Weak Supervision✓ Link14.86openai/whisper-large-v32023-11-06
Voxtral✓ Link15.04mistralai/Voxtral-Mini-3B-25072025-07-01
Moonshine: Speech Recognition for Live Transcription and Voice Commands✓ Link15.15usefulsensors/moonshine-base2024-10-21
Robust Speech Recognition via Large-Scale Weak Supervision✓ Link15.16openai/whisper-large-v3-turbo2024-10-01
OWSM-CTC: An Open Encoder-Only Speech Foundation Model for Speech Recognition, Translation, and Language Identification✓ Link15.18espnet/owsm_ctc_v3.1_1B2024-02-23
Conformer: Convolution-augmented Transformer for Speech Recognition✓ Link15.2nvidia/stt_en_conformer_ctc_large2022-04-09
Niagara-38m Sets a New Benchmark for Edge Speech Recognition15.46abr-ai/niagara-38m-batch.en2026-04-15
VibeVoice-ASR Technical Report✓ Link15.48microsoft/VibeVoice-ASR-HF2026-03-02
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link15.56nvidia/parakeet-ctc-0.6b2023-12-28
LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation✓ Link15.61efficient-speech/lite-whisper-large-v32025-02-26
LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation✓ Link15.61efficient-speech/lite-whisper-large-v3-turbo-acc2025-02-26
GLM-ASR-Nano: A robust, open-source speech recognition model✓ Link15.85zai-org/GLM-ASR-Nano-25122025-12-09
Voxtral Realtime✓ Link15.91mistralai/Voxtral-Mini-4B-Realtime-26022026-02-11
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link15.96nvidia/parakeet-rnnt-1.1b2023-12-27
On the Effects of Heterogeneous Data Sources on Speech-to-Text Foundation Models✓ Link15.96espnet/owsm_ctc_v3.2_ft_1B2024-09-24
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link16.47nvidia/parakeet-rnnt-0.6b2023-12-28
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link17.72nvidia/stt_en_fastconformer_ctc_large2023-06-08
Moonshine v2: Ergodic Streaming Encoder ASR for Latency-Critical Speech Applications✓ Link17.95usefulsensors/moonshine-streaming-tiny2026-02-12
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link18.01nvidia/stt_en_fastconformer_transducer_large2023-06-08
Niagara-38m Sets a New Benchmark for Edge Speech Recognition18.28abr-ai/niagara-19m-batch.en2026-04-15
Loquacious Set: 25,000 Hours of Transcribed and Diverse English Speech Recognition Data for Research and Commercial Use✓ Link18.54speechbrain/asr-conformer-largescaleasr2025-02-06
LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation✓ Link18.69efficient-speech/lite-whisper-large-v3-fast2025-02-26
Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languages✓ Link19.45facebook/omniASR-LLM-7B-v22025-12-12
Conformer: Convolution-augmented Transformer for Speech Recognition✓ Link19.65nvidia/stt_en_conformer_transducer_small2022-06-01
Conformer: Convolution-augmented Transformer for Speech Recognition✓ Link19.91nvidia/stt_en_conformer_ctc_small2023-06-12
Moonshine: Speech Recognition for Live Transcription and Voice Commands✓ Link20.42usefulsensors/moonshine-tiny2024-10-21
Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languages✓ Link22.1facebook/omniASR-CTC-7B-v22025-12-12
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations✓ Link29.45facebook/wav2vec2-large-960h-lv60-self2020-06-20
Robust wav2vec 2.0: Analyzing Domain Shift in Self-Supervised Pre-Training✓ Link30.44facebook/wav2vec2-large-robust-ft-libri-960h2021-04-02
HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units✓ Link31.53facebook/hubert-xlarge-ls960-ft2022-03-02
SpeechBrain: A General-Purpose Speech Toolkit✓ Link31.56speechbrain/asr-wav2vec2-librispeech2022-06-05
HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units✓ Link32.32facebook/hubert-large-ls960-ft2022-03-02
data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language✓ Link33.22facebook/data2vec-audio-large-960h2022-04-02
Scaling Speech Technology to 1,000+ Languages✓ Link35.14facebook/mms-1b-all2023-05-27
fairseq S2T: Fast Speech-to-Text Modeling with fairseq✓ Link35.36facebook/wav2vec2-conformer-rel-pos-large-960h-ft2022-04-18
fairseq S2T: Fast Speech-to-Text Modeling with fairseq✓ Link35.4facebook/wav2vec2-conformer-rope-large-960h-ft2022-04-18
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations✓ Link35.72facebook/wav2vec2-large-960h2020-06-20
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations✓ Link38.98facebook/wav2vec2-base-960h2022-03-02
data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language✓ Link41.11facebook/data2vec-audio-base-960h2022-03-02
Check out how to contribute new results. Then edit this file.