OpenCodePapers

speech-recognition-on-voxpopuli

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
Azure Speech at Build 2026: Powering Voice Agents with Real-Time and Life-like Experiences5.07microsoft/azure-speech-06-20262026-06-04
Introducing Zoom AI Services5.08zoom/scribe_v12026-03-10
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link5.12nvidia/parakeet-rnnt-1.1b2023-12-27
Efficient Sequence Transduction by Jointly Predicting Tokens and Durations✓ Link5.22nvidia/parakeet-tdt-1.1b2024-01-25
Training and Inference Efficiency of Encoder-Decoder Speech Models✓ Link5.38nvidia/canary-1b-flash2025-03-07
✓ Link5.38nvidia/canary-qwen-2.5b2025-07-17
✓ Link5.4ibm-granite/granite-speech-4.1-2b2026-04-29
Granite-speech: open-source speech-aware LLMs with strong English ASR capabilities✓ Link5.44ibm-granite/granite-speech-3.3-8b2025-06-19
5.46OpenMOSS-Team/MOSS-Transcribe-preview-2B2026-06-26
Data-Efficient On-Policy Distillation for Automatic Speech Recognition✓ Link5.47AutoArk-AI/ARK-ASR-3B2026-06-22
✓ Link5.51ibm-granite/granite-4.0-1b-speech2026-03-06
Less is More: Accurate Speech Recognition & Translation without Web-Scale Data✓ Link5.52nvidia/canary-1b2024-02-07
Introducing Cohere-transcribe: state-of-the-art speech recognition5.58CohereLabs/cohere-transcribe-03-20262026-03-26
Boson AI Launches Higgs STT 3 Speech-to-Text Model✓ Link5.62bosonai/higgs-audio-v3-8b-stt-v22026-04-27
NLE: Non-autoregressive LLM-based ASR by Transcript Editing✓ Link5.63ibm-granite/granite-speech-4.1-2b-nar2026-04-01
Granite-speech: open-source speech-aware LLMs with strong English ASR capabilities✓ Link5.64ibm-granite/granite-speech-3.3-2b2025-06-19
Pulse STT: Fast, Accurate Speech-to-Text5.65smallestai/pulse2026-01-28
Efficient Sequence Transduction by Jointly Predicting Tokens and Durations✓ Link5.68nvidia/parakeet-tdt-0.6b-v22025-05-01
Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs5.77microsoft/Phi-4-multimodal-instruct2025-02-24
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link5.78nvidia/parakeet-rnnt-0.6b2023-12-28
Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST✓ Link5.89nvidia/parakeet-tdt-0.6b-v32025-08-14
Introducing Resonant-1 and Resonant-1-flash5.9reson8/resonant-1-flash2026-04-08
Canary-1B-v2 & Parakeet-TDT-0.6B-v3: Efficient and High-Performance Models for Multilingual ASR and AST✓ Link5.93nvidia/canary-1b-v22025-08-14
Boson AI Launches Higgs STT 3 Speech-to-Text Model✓ Link5.93bosonai/higgs-audio-v3-stt2026-03-18
Qwen3-ASR Technical Report✓ Link5.99Qwen/Qwen3-ASR-1.7B2026-01-28
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link6.04nvidia/stt_en_fastconformer_transducer_large2023-06-08
Introducing Resonant-1 and Resonant-1-flash6.04reson8/resonant-12026-04-08
Training and Inference Efficiency of Encoder-Decoder Speech Models✓ Link6.11nvidia/canary-180m-flash2025-03-11
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link6.14nvidia/stt_en_fastconformer_ctc_large2023-06-08
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link6.29nvidia/parakeet-ctc-1.1b2023-12-28
Data-Efficient On-Policy Distillation for Automatic Speech Recognition✓ Link6.29AutoArk-AI/ARK-ASR-0.6B2026-05-25
Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languages✓ Link6.39facebook/omniASR-LLM-7B-v22025-12-12
Streaming Sequence-to-Sequence Learning with Delayed Streams Modeling✓ Link6.45kyutai/stt-2.6b-en2025-06-06
Introducing Universal-3 Pro: A new class of speech language model optimized for Voice AI6.49assemblyai/universal-3-pro2026-02-03
Conformer: Convolution-augmented Transformer for Speech Recognition✓ Link6.52nvidia/stt_en_conformer_ctc_large2022-04-09
Zipformer: A faster and better encoder for automatic speech recognition✓ Link6.56soundsgoodai/Zipformer-transducer-XL-290M2026-05-12
Voxtral✓ Link6.6mistralai/Voxtral-Small-24B-25072025-07-15
Efficient Sequence Transduction by Jointly Predicting Tokens and Durations✓ Link6.63nvidia/parakeet-tdt_ctc-110m2024-09-17
Ursa 2: Elevating speech recognition across 50+ languages6.64speechmatics/enhanced2024-10-11
Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition✓ Link6.77nvidia/parakeet-ctc-0.6b2023-12-28
Voxtral✓ Link6.77mistralai/Voxtral-Mini-3B-25072025-07-01
Qwen3-ASR Technical Report✓ Link6.8Qwen/Qwen3-ASR-0.6B2026-01-28
Loquacious Set: 25,000 Hours of Transcribed and Diverse English Speech Recognition Data for Research and Commercial Use✓ Link6.88speechbrain/asr-conformer-largescaleasr2025-02-06
Avalon: ASR for Human–AI Interaction6.88aquavoice/avalon-v1-en2025-08-22
OWSM v4: Improving Open Whisper-Style Speech Models via Data Scaling and Cleaning✓ Link7.12espnet/owsm_ctc_v4_1B2025-01-16
GLM-ASR-Nano: A robust, open-source speech recognition model✓ Link7.17zai-org/GLM-ASR-Nano-25122025-12-09
LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation✓ Link7.32efficient-speech/lite-whisper-large-v32025-02-26
LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation✓ Link7.44efficient-speech/lite-whisper-large-v3-fast2025-02-26
Conformer: Convolution-augmented Transformer for Speech Recognition✓ Link7.67nvidia/stt_en_conformer_transducer_small2022-06-01
Robust Knowledge Distillation via Large-Scale Pseudo Labelling✓ Link7.69distil-whisper/distil-large-v3.52024-12-05
On the Effects of Heterogeneous Data Sources on Speech-to-Text Foundation Models✓ Link7.75espnet/owsm_ctc_v3.2_ft_1B2024-09-24
LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation✓ Link7.77efficient-speech/lite-whisper-large-v3-acc2025-02-26
LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation✓ Link7.81efficient-speech/lite-whisper-large-v3-turbo-acc2025-02-26
Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languages✓ Link7.82facebook/omniASR-CTC-7B-v22025-12-12
VibeVoice-ASR Technical Report✓ Link7.9microsoft/VibeVoice-ASR-HF2026-03-02
Voxtral Realtime✓ Link8mistralai/Voxtral-Mini-4B-Realtime-26022026-02-11
Moonshine v2: Ergodic Streaming Encoder ASR for Latency-Critical Speech Applications✓ Link8.01usefulsensors/moonshine-streaming-medium2026-02-12
OWSM-CTC: An Open Encoder-Only Speech Foundation Model for Speech Recognition, Translation, and Language Identification✓ Link8.11espnet/owsm_ctc_v3.1_1B2024-02-23
CrisperWhisper: Accurate Timestamps on Verbatim Speech Transcriptions✓ Link8.6nyrahealth/CrisperWhisper2024-08-29
Niagara-38m Sets a New Benchmark for Edge Speech Recognition8.67abr-ai/niagara-38m-batch.en2026-04-15
Robust Speech Recognition via Large-Scale Weak Supervision✓ Link8.7openai/whisper-large-v32023-11-06
Conformer: Convolution-augmented Transformer for Speech Recognition✓ Link8.75nvidia/stt_en_conformer_ctc_small2023-06-12
Moonshine v2: Ergodic Streaming Encoder ASR for Latency-Critical Speech Applications✓ Link9.26usefulsensors/moonshine-streaming-small2026-02-12
Scaling Speech Technology to 1,000+ Languages✓ Link9.29facebook/mms-1b-all2023-05-27
Niagara-38m Sets a New Benchmark for Edge Speech Recognition9.82abr-ai/niagara-19m-batch.en2026-04-15
Moonshine: Speech Recognition for Live Transcription and Voice Commands✓ Link10.32usefulsensors/moonshine-base2024-10-21
Robust Speech Recognition via Large-Scale Weak Supervision✓ Link11.22openai/whisper-large-v3-turbo2024-10-01
fairseq S2T: Fast Speech-to-Text Modeling with fairseq✓ Link12.07facebook/wav2vec2-conformer-rope-large-960h-ft2022-04-18
fairseq S2T: Fast Speech-to-Text Modeling with fairseq✓ Link12.22facebook/wav2vec2-conformer-rel-pos-large-960h-ft2022-04-18
data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language✓ Link12.29facebook/data2vec-audio-large-960h2022-04-02
Moonshine: Speech Recognition for Live Transcription and Voice Commands✓ Link12.84usefulsensors/moonshine-tiny2024-10-21
Moonshine v2: Ergodic Streaming Encoder ASR for Latency-Critical Speech Applications✓ Link12.97usefulsensors/moonshine-streaming-tiny2026-02-12
SpeechBrain: A General-Purpose Speech Toolkit✓ Link13.02speechbrain/asr-wav2vec2-librispeech2022-06-05
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations✓ Link13.88facebook/wav2vec2-large-960h-lv60-self2020-06-20
HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units✓ Link14.99facebook/hubert-xlarge-ls960-ft2022-03-02
HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units✓ Link15.36facebook/hubert-large-ls960-ft2022-03-02
Robust wav2vec 2.0: Analyzing Domain Shift in Self-Supervised Pre-Training✓ Link16.22facebook/wav2vec2-large-robust-ft-libri-960h2021-04-02
data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language✓ Link17.85facebook/data2vec-audio-base-960h2022-03-02
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations✓ Link21.2facebook/wav2vec2-large-960h2020-06-20
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations✓ Link24.16facebook/wav2vec2-base-960h2022-03-02
Check out how to contribute new results. Then edit this file.