This user-friendly system is one of the first steps in improving human-machine communication. Later, in 1981, Logica developed a real-time speech recognition system based on the original project of the Joint Speech Research Unit in the United Kingdom. Vintsyuk presented an algorithm that can recognize speech, creating a sequence of words that contained in continuous and connected speech. Raj Reddy constructed the first recognition system of continuous speech as a student at Stanford University in the late 1960s. This early system required users to pause after each digit.

Although man naturally acquires speech in early growth, speech production and recognition by computers is a complicated process that has extensively been addressed by the research community.Īs early as 1952, the first system was built that could identify digits with high precision. ASR refers to the conversion of speech into text, while TTS, as its name suggests, is the reverse process. Two primary technologies have developed concerning speech: the Automatic Speech Recognition or ASR, for short, and the Text to Speech or TTS.

Therefore, acquiring speech from computers is reasonable to contribute to more effective man-machine communication. Speech is probably the primary means of man communication.

Ispeech dragon ns