90 days of the enterprise, pro and team plans free, no card needed. See the plans
LLMs1 min read
Hugging Face introduced the ASRA benchmark to evaluate and compare the optimization of speech recognition models. The benchmark focuses on measuring the impact of various optimization techniques on model performance.
By OpenSmartRoute editorial · written through the router by writer-small
From Hugging Face blog - “Measuring benchmark optimization in speech recognition”
The ASRA benchmark provides a standardized method for assessing speech recognition models. It measures performance across a range of metrics, including accuracy and latency. The benchmark includes a suite of models and optimization techniques for comparison. The benchmark is designed to facilitate the selection of models best suited for specific applications.
Source: https://huggingface.co/blog/asr-benchmark-optimization
Keep reading
Agents6 min read
Google released TorchTPU, a native PyTorch backend for Cloud TPUs, and updated TorchTitan to train Llama 3, Qwen 3 and FLUX models on TPU with near-JAX performance.
LLMs8 min read
Google released ML Drift, an open-source GPU engine for on-device AI. It replaces the legacy TensorFlow Lite GPU delegate and runs on diverse hardware.
Daily in your inbox
New models, LLM releases, agent frameworks and research - the day's posts in one email.
AI16 min read
Block released Dessa, an AI model trained solely on human speech to detect emotions. It processes audio directly without converting it to text first.
