TAAFT
Free mode
100% free
Freemium
Free Trial
Deals
Companies
NVIDIA
Models
Canary 1B v2

Canary 1B v2

By NVIDIA
Canary 1B v2 is a multilingual, multitask automatic speech recognition model in NVIDIA’s Canary line, built to transcribe speech across a broad set of languages (the model documentation lists 25) and intended for robust speech-to-text performance across varied acoustic conditions, with components and configuration aligned with NVIDIA NeMo ASR model design.
New Audio Gen 2
Released: August 14, 2025

Overview

Multilingual, multitask speech recognition model (ASR) in the NVIDIA NeMo family.

🗒Transcription

About NVIDIA

Industry: Computer Hardware Manufacturing
Company Size: 36000
Location: Santa Clara, California, US
Website: nvidia.com
View Company Profile

Tools using Canary 1B v2

No tools found for this model yet.

Related Models

  • Gen 4
    Cabina transcriber v1
    Cabina transcriber v1 is a real-time speech-to-text model with stable timestamps, punctuation, and optional speaker labels for meetings, calls, and streams.
    🗒Transcription
    Audio
    Released 7mo ago
  • Gen 4
    Suno V5
    Suno v5 is a text-to-music model that turns prompts into full songs with clearer vocals, tighter rhythm, stronger style control, and better long-form structure. It supports continuation, reference guidance, and quick iteration from demo to release-ready tracks.
    🎵Music🎵Music lyrics
    Audio
    Released 5mo ago
  • Gen 4
    Suno Bark
    Suno Bark is an open-source text-to-audio model that generates expressive speech—and even music, effects, and nonverbal cues—from plain text. It’s multilingual, handles emotion and prosody, and can mimic styles with short reference prompts for rapid voiceovers and sound design.
    🔊Text to speech🎵Music
    Audio
    Released 1y ago
Last updated: February 25, 2026
0 AIs selected
Clear selection
#
Name
Task