Speak&Find Recent Changes

Recent changes in Version 1.3 (compared to 1.2):

• ConText Management: Speak&Find supports topic specific vocabularies to increase the accuracy. You can install any Philips certified ConText or simply use the default general ConText.

• Speaker Training: If the speaker is known, you can create and train acoustic models for specific speakers. By using this feature you can increase the accuracy for trained speakers significantly.

• Recognition Data Storage: Any data used by Speak&Find is now stored via the data storage which can be implemented by you. This allows you to use any storage mechanism you like (e.g. flat files or databases)

• 'Online Setting' Sensitivity: If you like to change the sensitivity of Speak&Find you don't need to re-initialize the engine any longer.

• 'Online Setting' Accuracy: If you like to change the accuracy of Speak&Find you don't need to re-initialize the engine any longer.