ARM-software/ML-KWS-for-MCU
Keyword spotting on Arm Cortex-M Microcontrollers
Implements multiple neural network architectures (DNN, CNN, LSTM, GRU, CRNN, DS-CNN) trained on speech commands with TensorFlow, optimized for quantization and deployed on Cortex-M boards via frozen `.pb` models. Provides end-to-end training, evaluation, and inference pipelines with configurable layer dimensions, plus pretrained models with published accuracy/memory/operation metrics for direct deployment.
1,231 stars. No commits in the last 6 months.
Stars
1,231
Forks
427
Language
C
License
Apache-2.0
Category
Last pushed
Apr 10, 2019
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ARM-software/ML-KWS-for-MCU"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
Ant-Brain/EfficientWord-Net
OneShot Learning-based hotword detection.
Tony607/Keras-Trigger-Word
How to do Real Time Trigger Word Detection with Keras | DLology
ardamavi/Vocalization-Sign-Language-iOS
Vocalization sign language iOS App with deep learning using CoreML.
roman-vygon/triplet_loss_kws
Learning Efficient Representations for Keyword Spotting with Triplet Loss
Adamiito0909/mlx-swift-audio
🎤 Enhance your apps with MLX Swift Audio, offering robust text-to-speech and speech-to-text...