System and method for standardized speech recognition infrastructure

A system comprising:

• a processor;
• and
• a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
• receiving speech from a user on a device;
• and
• communicating the speech to a separate device and away from the device, wherein the separate device:
• determines to apply one of supervised training and unsupervised training;
• and
• when supervised training is selected:
• determines whether available data are sufficient to build a new speech recognition model;
• when the available data is sufficient to build the new speech recognition model, builds the new speech recognition model using the available data;
• when the available data is not sufficient to build the new speech recognition model:
• selects an existing speech recognition model;
• and
• generates an adapted speech recognition model based on transformations generated from the existing speech recognition model based on the speech and associated transcriptions.

A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:

• receiving speech from a user on a device;
• and
• communicating the speech to a separate device and away from the device, wherein the separate device:
• determines to apply one of supervised training and unsupervised training;
• and
• when supervised training is selected:
• determines whether available data are sufficient to build a new speech recognition model;
• when the available data is sufficient to build the new speech recognition model, builds the new speech recognition model using the available data;
• when the available data is not sufficient to build the new speech recognition model:
• selects an existing speech recognition model;
• and
• generates an adapted speech recognition model based on transformations generated from the existing speech recognition model based on the speech and associated transcriptions.