Tag

#ASR

6 articles

Meet S1-mini: Superwhisper’s 462 MB Open-Weights Text Normalizer That Turns Raw ASR Transcripts Into Clean Written Text

Learn how to use S1-mini, a 462 MB open-weight text normalizer that cleans up raw ASR transcripts by removing fillers and self-corrections.

Aug 2018

Best Open Speech Recognition (ASR) Models in 2026: WER, Languages, Latency, and License Compared

In 2026, the open-source speech recognition landscape has diversified beyond Whisper's dominance, with several models now competing closely on performance metrics. A detailed comparison reveals nuanced trade-offs in accuracy, language support, and latency.

Jul 2345

Interfaze Ships diffusion-gemma-asr-small, an Open-Source Diffusion ASR Model Transcribing Six Languages via DiffusionGemma’s Parallel Denoising Decoder

Learn how a new AI model called diffusion-gemma-asr-small uses a 'diffusion' approach to transcribe speech in six languages more efficiently than traditional methods.

Jul 265

tech

How to Use NVIDIA Canary-1B-v2 for ASR, Translation, and Automatic SRT Subtitle Export in Python

NVIDIA's Canary-1B-v2 model enables developers to build multilingual ASR and translation pipelines with automatic SRT subtitle export, showcasing advancements in AI-powered speech processing.

Jun 2369

IBM Releases Two Granite Speech 4.1 2B Models: Autoregressive ASR with Translation and Non-Autoregressive Editing for Fast Inference

IBM has launched two new Granite Speech 4.1 2B models — one autoregressive for high-accuracy speech recognition with translation, and one non-autoregressive for fast inference.

Apr 30102

Cohere AI Releases Cohere Transcribe: A SOTA Automatic Speech Recognition (ASR) Model Powering Enterprise Speech Intelligence

Cohere AI has released Cohere Transcribe, a state-of-the-art automatic speech recognition model designed to transform audio into actionable text for enterprise use cases.

Mar 26127