How to Use NVIDIA Canary-1B-v2 for ASR, Translation, and Automatic SRT Subtitle Export in Python
NVIDIA Canary-1B-v2 now does multilingual ASR, real-time translation, and auto-subtitle generation in a single pipeline.
Why it matters
Demonstrates practical deployment of NVIDIA's speech foundation model for production workflows—relevant for teams building voice-first AI applications and reducing transcription infrastructure costs.
The key facts
12 to knowNVIDIA Canary-1B-v2 supports multilingual ASR and speech translation
Supports English ASR with translation to French, German, Spanish, Italian
Enables automatic SRT subtitle export with word-level and segment timestamps
Inference benchmarked on GPU-enabled runtime
Batch processing and long-form transcription tested
Published June 23, 2026
NVIDIA Canary-1B-v2 supports ASR and multilingual translation (French, German, Spanish, Italian)
Model handles end-to-end pipeline: audio input → transcription → translation → SRT subtitle export
Supports long-form transcription and batch processing
Inference benchmarking available for performance planning
16 kHz mono audio preprocessing required
GPU-accelerated runtime for deployment
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: In this tutorial, we build a multilingual ASR and speech translation pipeline with NVIDIA Canary-1B-v2. We load the model on a GPU-enabled runtime, prepare audio into 16 kHz mono, and run English ASR. We then translate speech into French, German, Spanish, and Italian, and extract word and segment…
