search

kahrendt / microWakeWord

A TensorFlow based wake word detection training framework using synthetic sample generation suitable for certain microcontrollers.
Apache License 2.0
276 stars 20 forks source link

init features_offset variable for cases where data_length > features_… #20

Closed alior101 closed 5 months ago

alior101 commented 5 months ago

when training with datasets with truncation set to "split" and when the datalength is greater than the featuers length configured, the features_offset variable is left un-initialized and is later used un-initialized which crashes the script. This commit fixes that...