HF model

reazon-research / reazonspeech-nemo-v2

reazonspeech-nemo-v2 is a 619‑M‑parameter Japanese long‑form ASR model built on an improved Fast‑Conformer with Linearly Scalable Attention, trained on the ReazonSpeech v2.0 corpus, offering multi‑hour inference via a subword RNN‑T decoder (3000‑token SentencePiece) and distributed under Apache 2.0.

Description in other languages

Statistics

Statistics as of Nov 16, 2026

Hugging Face downloads (last 30 days)
4,000
Hugging Face likes
41
Listed
May 13, 2024

Downloads (30-day window) over time

010k20k30kNovSepJulMayMarJan 20264,000
Show as a table
MonthDL (30d)
Dec 20257,000
Jan 20266,000
Feb 20264,000
Mar 202610,000
Apr 202622,000
May 20261,000
Jun 20261,000
Jul 20261,000
Aug 20261,000
Sep 20262,000
Oct 20263,000
Nov 20264,000

Downloads over the 30 days before the last snapshot of each month.

Popular in the same category

Resources in similar research fields

Learn from slides & articles