Chapter 4c - Voice User Interface (VUI)
Chapter 4c - Voice User Interface (VUI)
• https://www.canakit.com/raspberry-pi-zero-wireless.html
3.2 Raspberry Pi
based Voice
application
architecture
3.3 ALEXA-
AVS – Sample
app
1. https://github.com/alexa
/alexa-avs-sample-app
2. https://www.codemade.i
o/project-raspberry-pi-al
exa-voice-service/
4. Voice Interactions
Types
1. Command based
2. Speech to Text (STT) / Text to
Speech (TTS)
3. Conversational
4. Personal identification – Biometric
4.1 Command Based
4.2 Speech to Text: Dictation
4.3 Text to Speech: Language translation
• Google Text to Speech API - Powered by Machine Learning:
https://www.google.com/aclk?sa=l&ai=DChcSEwjen9Ldh97tAhWLMCsKHaa0DR0YABA
DGgJzZg&sig=AOD64_2p9x0QerqXgkmAahEE8BxM0VkNjw&q&adurl&ved=2ahUKEw
i3ycvdh97tAhXOgtgFHa9SAWoQ0Qx6BAgHEAE
4.4 Conversational / AI based: chatbot
4.5 Personal Identification: Voice Biometrics
5. VUI – Speech Recognition Technology
• Allows people to interact with a computer, smartphone or other
device through voice commands
• The prime examples
Apple’s Siri
Amazon’s Alexa
Google’s Assistant
Microsoft’s Cortana
• Goal: To communicate
naturally with machine
• AI powered & voice driven:
Tone of voice
Choice of words etc.
9. Steps for designing VUI
i. Research audience
ii. Define capabilities and shape the product
Creating key scenarios of interaction
Make sure these scenarios work with voice
The three factors- Intent, Utterance, Slot