Views
No views yet
[!Note] This model is the quantized INT4 ONNX version of nvidia/nemotron-3.5-asr-streaming-0.6b, adding language-ID prompt conditioning to support transcription across 40 language-locales from a single model. It supports streaming inference with 0.56 seconds of latency and is simshipped alongside the baseline NVIDIA model.
| Tier | Languages (locales) |
|---|---|
| Transcription-ready (19 locales) | English (en-US, en-GB), Spanish (es-US, es-ES), French (fr-FR, fr-CA), Italian (it-IT), Portuguese (pt-BR, pt-PT), Dutch (nl-NL), German (de-DE), Turkish (tr-TR), Russian (ru-RU), Arabic (ar-AR), Hindi (hi-IN), Japanese (ja-JP), Korean (ko-KR), Vietnamese (vi-VN), Ukrainian (uk-UA) |
| Broad-coverage (13 locales) | Polish (pl-PL), Swedish (sv-SE), Czech (cs-CZ), Norwegian Bokmål (nb-NO), Danish (da-DK), Bulgarian (bg-BG), Finnish (fi-FI), Croatian (hr-HR), Slovak (sk-SK), Mandarin (zh-CN), Hungarian (hu-HU), Romanian (ro-RO), Estonian (et-EE) |
| Adaptation-ready (8 locales) | Greek (el-GR), Lithuanian (lt-LT), Latvian (lv-LV), Maltese (mt-MT), Slovenian (sl-SI), Hebrew (he-IL), Thai (th-TH), Norwegian Nynorsk (nn-NO) |
Note: Transcription-ready and broad-coverage locales (32 total) produce ASR transcription out of the box; adaptation-ready locales require fine-tuning on in-domain data to enable full transcription. The model supports uppercase and lowercase letters, punctuation, spaces, and apostrophes.
| # | Lang | Metric | FLEURS | CV | MLS | VoxPopuli |
|---|---|---|---|---|---|---|
| 1 | ko (ko_kr) | CER | 6.19 | – | – | – |
| 2 | it (it_it) | WER | 6.69 | – | – | – |
| 3 | es (es_419) | WER | 7.28 | 9.16 | 6.07 | 9.96 |
| 4 | pt (pt_br) | WER | 9.05 | 12.08 | 8.85 | – |
| 5 | hi (hi_in) | WER | 9.79 | – | – | – |
| 6 | de (de_de) | WER | 12.25 | 12.80 | 9.91 | 16.90 |
| 7 | ar (ar_eg) | WER | 12.31 | – | – | – |
| 8 | fr (fr_fr) | WER | 12.55 | 14.38 | 8.61 | 11.99 |
| 9 | ru (ru_ru) | WER | 13.52 | – | – | – |
| 10 | vi (vi_vn) | WER | 14.45 | – | – | – |
| 11 | ja (ja_jp) | CER | 15.12 | – | – | – |
| 12 | nl (nl_nl) | WER | 16.27 | 14.29 | 18.73 | 28.89 |
| 13 | tr (tr_tr) | WER | 16.69 | – | – | – |
| 14 | uk (uk_ua) | WER | 18.16 | – | – | – |
| 15 | pl (pl_pl) | WER | 21.53 | 26.85 | 16.31 | 35.35 |
| 16 | zh (cmn_hans_cn) | CER | 22.88 | – | – | – |
| 17 | sk (sk_sk) | WER | 23.54 | – | – | – |
| 18 | nb (nb_no) | WER | 23.49 | – | – | – |
| 19 | cs (cs_cz) | WER | 25.66 | – | – | – |
| 20 | fi (fi_fi) | WER | 25.96 | – | – | – |
| 21 | bg (bg_bg) | WER | 26.35 | – | – | – |
| 22 | sv (sv_se) | WER | 28.94 | – | – | – |
| 23 | hr (hr_hr) | WER | 31.43 | – | – | 49.88 |
| 24 | da (da_dk) | WER | 33.78 | – | – | – |
| 25 | hu (hu_hu) | WER | 34.97 | – | – | – |
| 26 | ro (ro_ro) | WER | 35.58 | – | – | – |
| 27 | et (et_ee) | WER | 36.51 | – | – | – |
| 28 | el (el_gr) | WER | 48.25 | – | – | – |
| 29 | he (he_il) | WER | 51.24 | – | – | – |
| 30 | lt (lt_lt) | WER | 53.00 | – | – | – |
| 31 | sl (sl_si) | WER | 56.89 | – | – | – |
| 32 | mt (mt_mt) | WER | 62.49 | – | – | – |
| 33 | lv (lv_lv) | WER | 87.87 | – | – | – |
| 34 | th (th_th)* | CER | 100.00* | – | – | – |