Continuous Speech Recognition with Mid Speech interim using Google Speech API
******* My LinkedIn- / yohanrw *****
This is a sample application. It is using Google Speech API and developed for Android. It demonstrates the continuous speech recognition, getting mid speech interim and also identifying pre-defined words.
Continuous recognition is considered as "Impossible" and google also says the same, but in this app it is achieved.
In the Mid Speech interims, you get lot of duplicated results. For an example, if you speak "I am Yohan", then the mid speech results will be like below.
1st result - I
2nd resilt - I am
3rd result- Bohan
4th result - I am Yohan
When it is printed, it will look like "I I am Bohan I am Yohan"
This duplication is removed from this application using a proven scientific formula.