{"v":1,"id":"model:hf:nvidia/nemotron-speech-streaming-en-0.6b","slug":"model-nvidia-nemotron-speech-streaming-en-0-6b","kind":"model","category":"speech","title":"nemotron-speech-streaming-en-0.6b","summary":"June 4, 2026: New multilingual model released: NVIDIA Nemotron 3.5 ASR Streaming 0.6B extends this English streaming ASR model to 40 language-locales in a single 600M-parameter model. It supports lan…","source":{"provider":"hf","ref":"nvidia/nemotron-speech-streaming-en-0.6b","url":"https://huggingface.co/nvidia/nemotron-speech-streaming-en-0.6b","rev":"ebe59e5a817142986528bbbee5dba8db7b38ed50","fetchedAt":"2026-10-02T20:52:17.292Z","etag":"W/\"2823-XI0v4zVFRiByI7hf58s8ZBBeXRk\""},"author":{"name":"nvidia","url":"https://huggingface.co/nvidia"},"license":{"spdx":null,"raw":"other","url":"https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/","open":null,"note":"custom license: read it at the source before installing"},"metrics":{"downloads":331849,"likes":632,"stars":18539,"openIssues":325,"lastRelease":{"tag":"v3.0.0","at":"2026-08-07T00:13:21Z"},"pushedAt":"2026-10-02T16:11:38Z","takenAt":"2026-10-02T20:52:17.292Z"},"tags":["nemo","safetensors","gguf","nemotron_asr_streaming","feature-extraction","transformers","speech-recognition","cache-aware ASR","automatic-speech-recognition","streaming-asr","speech","audio","FastConformer","RNNT","Parakeet","ASR","pytorch","NeMo","hf-asr-leaderboard","dataset:nvidia/Granary","dataset:YTC","dataset:Yodas2","dataset:LibriLight","dataset:librispeech_asr","dataset:fisher_corpus","dataset:Switchboard-1","dataset:WSJ-0","dataset:WSJ-1","dataset:National-Singapore-Corpus-Part-1","dataset:National-Singapore-Corpus-Part-6","dataset:vctk","dataset:voxpopuli","dataset:europarl","dataset:multilingual_librispeech","dataset:fleurs","dataset:mozilla-foundation/common_voice_8_0","dataset:MLCommons/peoples_speech","dataset:google/speech_commands","model-index"],"pipeline":"automatic-speech-recognition","links":{"github":"NVIDIA-NeMo/Speech"},"updatedAt":"2026-08-05T21:45:11.000Z","collectedAt":"2026-10-02T20:52:17.292Z","review":{"numbers":["331,849 downloads on Hugging Face","632 likes","license other","0.7 GB for nemotron-speech-streaming-en-0.6b.q8_0.gguf","18,539 stars on NVIDIA-NeMo/Speech","325 open issues and PRs","last release v3.0.0 on 2026-08-07"],"log":null},"trust":"unlabeled","health":{"status":"alive","checkedAt":"2026-10-02T20:52:17.292Z","http":200},"description":"# Nemotron ASR Streaming\n\nh1, h2, h3, h4, h5, h6 {\n  color: #76b900; /* NVIDIA green */\n  font-weight: 700;\n}\n\nhr {\n  border: none;\n  border-top: 1px solid #e5e7eb;\n  margin: 2rem 0;\n}\n\n/* Improve list spacing */\nul, ol {\n  margin-top: 0.5rem;\n  margin-bottom: 0.5rem;\n}\n\n/* Badge alignment consistency */\nimg {\n  display: inline;\n  vertical-align: middle;\n}\n\n[](#model-architecture)\n| [](#model-architecture)\n| [](#datasets)\n\n> [!Note]\n> June 4, 2026: **New multilingual model released:** [NVIDIA Nemotron 3.5 ASR Streaming 0.6B](https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b) extends this English streaming ASR model to **40 language-locales** in a single 600M-parameter model. It supports language-ID prompt conditioning, optional automatic language detection, punctuation and capitalization, and configurable low-latency streaming chunk sizes.\n>\n> March 12, 2026: nemotron-asr-streaming was released with updated checkpoint (trained on larger corpora). For the older checkpoint released in January 2026, please refer to the [nemotron-speech-streaming-jan2026](https://huggingface.co/nvidia/nemotron-speech-streaming-en-0.6b/tree/nemotron-speech-streaming-jan2026) branch.\n\nNemotron-ASR-Streaming is an English, streaming Automatic Speech Recognition (ASR) engineered to deliver high-quality English transcription across both low-latency streaming and high-throughput batch workloads. Developed by NVIDIA, this 600M parameter model transcribes speech into text with native support for punctuation and capitalization and offers runtime flexibility with configurable chunk sizes, including 80ms, 160ms, 560ms, and 1120ms.\n\nBy leveraging the state-of-the-art Cache-Aware FastConformer-RNNT architecture, the model eliminates redundant overlapping computations common in traditional \"buffered\" streaming. This allows it to process only new audio chunks while reusing cached encoder context, significantly improv…\n\nSource: https://huggingface.co/nvidia/nemotron-speech-streaming-en-0.6b","install":{"kind":"model","hfId":"nvidia/nemotron-speech-streaming-en-0.6b","gated":false,"format":"gguf","files":[{"name":"model.safetensors","size":2472413604,"sha256":"bddd8a7300826efd19cf7e01f1c7db8402bed6786fc4c7739632894f69c71473"},{"name":"nemotron-speech-streaming-en-0.6b.q8_0.gguf","size":699872960,"quant":"Q8_0","sha256":"d9a01898d2a611c8764e23a1c2f45e70bbd5a425dc4de93692ac951dd603812d"}],"totalBytes":3172286564,"suggestedFile":"nemotron-speech-streaming-en-0.6b.q8_0.gguf","requirements":{"ramGb":2,"diskBytes":699872960,"note":"estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models"},"runWith":["whisper.cpp"],"command":"lsh models install hf:nvidia/nemotron-speech-streaming-en-0.6b"}}