.. _segmentation_api:

Segmentation
============

Segmentation aims to break long audio files into chunks of speech.

.. note::

   The current implementation of segmentation uses only Voice Activity Detection (VAD) features. There's been some work towards getting a full speaker diarization set up going with :ref:`training_ivector_api` but that's largely planned for 2.1.

.. toctree::

   main
   helper