NVIDIA releases open-source Nemotron 3.5 Transcribe and Granary v2 for multilingual transcription
Current overview (not historical)
NVIDIA has released two open-source AI tools for multilingual transcription: Nemotron 3.5 Transcribe and Granary v2. Nemotron 3.5 Transcribe supports automatic speech recognition for 40 languages, translation into 31 languages, and speaker diarization for 13 languages, and can run on a single GPU. Granary v2 generates verbatim transcripts, cleaned scripts, and stylized transcriptions for each audio file, allowing developers to request consistent styling via prompts. These releases expand NVIDIA's open-source AI offerings in voice technology.
Written with AI from the reporting · updated 10/28/2026
EventGraph Explorer
One time cutoff for reports, relations and evidence.
New at this step ·Nvidia releases open Nemotron 3.5 Transcribe model combining transcription, translation and speaker recognition
- Reports in replay
- 3
- First-party
- 1
- Sources
- 3
Reports first seen at this step · 1
Discovery-time replay; the linked graph and evidence share this cutoff. The full overview and activity remain current.
Event developments
1 developments10/28/2026, 11:07:27 PM
NVIDIA Nemotron 3.5 Transcribe adds multilingual transcription, speaker attribution and audio Q&A as open modelHugging Face Blog
Filtered by first discovery; only representative reports already discovered by this cutoff are shown. Story grouping may have changed later, so this is not a complete historical snapshot.
Relations at this time
Loading time-scoped relations…
Bounded reconstruction, not an exact snapshot: entity names, relationship text and story grouping may include later revisions. Overview and full source archive below remain current.
Story route
A station-by-station route from the first public evidence onward; every station carries its own evidence and uncertainty.
- 1Oct 28, 00:00recorded Oct 28, 23:07revised
NVIDIA Nemotron 3.5 Transcribe and Granary v2 发布
Read original source ↗ Hugging Face BlogReplay this moment in the explorer
Uncertainty marks flag where to check the sources; they are not proven causation.
Full source archive (current)
Current public reports; outside the historical replay.
First-party sources · 1 reports
- Hugging Face Blog · Oct 28, 22:58NVIDIA Nemotron 3.5 Transcribe adds multilingual transcription, speaker attribution and audio Q&A as open model
NVIDIA Nemotron 3.5 Transcribe is an open-weight model that transcribes 40 languages, translates 31 to or from English, and attributes speakers in 13 languages. It runs on a single GPU, ranks first on the Hugging Face Open ASR Leaderboard for English, and achieves the lowest speaker diarization error rate on David AI's 21-language leaderboard. Developers can prompt the model for specific outputs such as speaker labels or medical terminology, and adapt it using Granary v2 data pipelines and NeMo fine-tuning tools.
Read original source ↗
Independent coverage · 2 reports
- The Decoder · Oct 28, 23:49Nvidia releases open Nemotron 3.5 Transcribe model combining transcription, translation and speaker recognition
Nvidia has released Nemotron 3.5 Transcribe, an open speech recognition model that processes text instructions to handle transcription, translation, and speaker identification in a single system. The model supports 40 languages for transcription and 31 for translation, and can identify speakers across 13 languages. It handles recordings up to an hour long with up to eight speakers. According to Nvidia, the model ranks first among open-weight models on Hugging Face's English Open ASR Leaderboard with a 5.33 percent average word error rate, and scores 2.81 percent on Artificial Analysis benchmarks, approaching top closed models while using significantly less compute. It also leads in speaker recognition, outperforming 14 tested systems on VoiceArena's benchmark.
Read original source ↗ - TechCrunch · AI · Oct 28, 23:00Nvidia releases Nemotron 3.5 Transcribe, a free open-weight model that transcribes audio in up to 40 languages
Nvidia released Nemotron 3.5 Transcribe, a free open-weight model that transcribes audio in up to 40 languages and can translate 31 languages to and from English. The model handles recordings up to an hour long and can distinguish up to eight speakers. Nvidia claims it outperforms rivals on benchmarks including Hugging Face's Open ASR Leaderboard. The model, with 30 billion parameters, can run on a single chip ranging from edge devices to datacenter chips. It is designed for developers building business meeting and medical transcription apps.
Read original source ↗
Different reports do not necessarily disagree. Check original sources and publication dates.
Report timeline
Explore every side through dated source reports.
- The DecoderNvidia releases open Nemotron 3.5 Transcribe model combining transcription, translation and speaker recognition
Nvidia has released Nemotron 3.5 Transcribe, an open speech recognition model that processes text instructions to handle transcription, translation, and speaker identification in a single system. The model supports 40 languages for transcription and 31 for translation, and can identify speakers across 13 languages. It handles recordings up to an hour long with up to eight speakers. According to Nvidia, the model ranks first among open-weight models on Hugging Face's English Open ASR Leaderboard with a 5.33 percent average word error rate, and scores 2.81 percent on Artificial Analysis benchmarks, approaching top closed models while using significantly less compute. It also leads in speaker recognition, outperforming 14 tested systems on VoiceArena's benchmark.
- TechCrunch · AINvidia releases Nemotron 3.5 Transcribe, a free open-weight model that transcribes audio in up to 40 languages
Nvidia released Nemotron 3.5 Transcribe, a free open-weight model that transcribes audio in up to 40 languages and can translate 31 languages to and from English. The model handles recordings up to an hour long and can distinguish up to eight speakers. Nvidia claims it outperforms rivals on benchmarks including Hugging Face's Open ASR Leaderboard. The model, with 30 billion parameters, can run on a single chip ranging from edge devices to datacenter chips. It is designed for developers building business meeting and medical transcription apps.
- Hugging Face BlogSelectedNVIDIA Nemotron 3.5 Transcribe adds multilingual transcription, speaker attribution and audio Q&A as open model
NVIDIA Nemotron 3.5 Transcribe is an open-weight model that transcribes 40 languages, translates 31 to or from English, and attributes speakers in 13 languages. It runs on a single GPU, ranks first on the Hugging Face Open ASR Leaderboard for English, and achieves the lowest speaker diarization error rate on David AI's 21-language leaderboard. Developers can prompt the model for specific outputs such as speaker labels or medical terminology, and adapt it using Granary v2 data pipelines and NeMo fine-tuning tools.
Event activity trend
- Latest comparable
- 3
- Comparable peak
- 2910月29日 00:00
- 24h change
- -83%
The chart compares consistently observed participants only. Hover or tap to inspect each hour; use arrow keys to move between observations.