Skip to main content

Framework for building deep neural network models for sound, speech, and voice AI

Project description

Sonus AI: Framework for simplified creation of deep NN models for sound, speech, and voice AI

Sonus AI includes functions for pre-processing training and validation data and creating performance metrics reports for key types of Keras models: - recurrent, convolutional, or a combination (i.e. RCNNs) - binary, multiclass single-label, multiclass multi-label, and regresssion - training with data augmentations: noise mixing, pitch and time stretch, etc.

Sonus AI python functions are used by:
  • Aaware Inc. sonusai executable: Easily create train/validation data, run prediction, evaluate model performance

  • Keras model scripts: User python scripts for keras model creation, training, and prediction. These can use sonusai-specific data but also some general useful utilities for trainining rnn-based models like CRNN’s, DSCRNN’s, etc. in Keras

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

sonusai-0.6.10.tar.gz (1.9 MB view hashes)

Uploaded Source

Built Distribution

sonusai-0.6.10-py3-none-any.whl (2.0 MB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page