PulseAugur
中
实时 06:21:58
English(EN) NVIDIA Releases Nemotron 3.5 Transcribe: An Open 30B MoE Model for 40-Language Transcription with ~4B Active Parameters

NVIDIA 发布 Nemotron 3.5 Transcribe,一款支持 40 种语言的开源语音大语言模型

NVIDIA 发布了 Nemotron 3.5 Transcribe,这是一款开源权重语音大语言模型,能够转录 40 种语言的音频。该模型拥有约 43 亿活跃参数和 65,536 个 token 的上下文窗口,可处理长达一小时的音频。它提供多种输出风格,包括逐字记录、书面形式、说话人归属、上下文和翻译文本,旨在将独立的 ASR、说话人分离和对齐工具整合为单一解决方案。虽然它在英语和多语言转录基准测试中领先于开源模型,但顶级的商业 API 仍然在性能上保持优势。 AI

影响 为开源语音转录模型设定了新的 SOTA(State-of-the-Art),可能减少对多个专用工具的依赖。

排序理由 NVIDIA 发布 Frontier-lab 模型,附带系统卡。[lever_c_降级自 frontier_release: ic=1 ai=1.0]

在 MarkTechPost 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

NVIDIA 发布 Nemotron 3.5 Transcribe,一款支持 40 种语言的开源语音大语言模型

本文如何被排名

Signal score
62 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Significant
NVIDIA 发布 Frontier-lab 模型,附带系统卡。[lever_c_降级自 frontier_release: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
model release, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

  1. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    NVIDIA 发布 Nemotron 3.5 Transcribe:一款支持 40 种语言转录的开源 30B MoE 模型,约有 4B 活跃参数

    <p>NVIDIA released Nemotron 3.5 Transcribe, an open-weights speech LLM with 4.3B active parameters. It transcribes 40 languages, adds speaker labels and word timestamps, and posts a 4.71% mean WER on the English Open ASR Leaderboard.</p> <p>The post <a href="https://www.marktechp…