CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

24
CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony

Transcript of CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

Page 1: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

CS378 - Mobile Computing

Speech to Text,Text to Speech,

Telephony

Page 2: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

2

SPEECH RECOGNITION - SPEECH TO TEXT

Page 3: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

3

Speech Recognition• android.speech package• Simplest example - start an Intent for a

resultRecognizerIntent.ACTION_RECOGNIZE_SPEECH

• uses network– true on the dev phones– doesn't work on emulator

Page 4: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

4

Speech Recognition

Page 5: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

5

Starting Intent

Page 6: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

6

Responding to Result• Note: list of results, ordered by confidence

Page 7: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

7

Results

Page 8: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

8

Modified Version• Show confidence scores

Page 9: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

9

Sample Results

Page 10: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

10

Sample Results

Page 11: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

11

Android SpeechRecognizer• Allows lower level control of SpeechRecognizer• If you don't want to use the intent and built in

activity• must call methods on the main activity thread• must have RECORD_AUDIO permission• likely uses a service– don't use for continuous recognition

• Requires RecognitionListener for call backs

Page 12: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

12

TEXT TO SPEECH

Page 13: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

13

Android TextToSpeech Class• Not as easy to use as SpeechToText• … but not too bad• In example, edit text to enter text and

button to launch speech to text object

Page 14: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

14

Text to Speech UI Example

Page 15: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

15

Text To Speech• Create Text To Speech Object• constructor requires a Context and a

TextToSpeech.OnInitListener

Page 16: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

16

TextToSpeech.OnInitListener• one method, onInit• passed an int for status

Page 17: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

17

onClick Method for Text to Speech• get text from EditText• pass to TextToSpeech object

• Demo• Other locales available–does not translate text

Page 18: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

18

Text To Speech• methods to:• save to a sound file• map text to sound files other than

default• check if languages available• set rate and pitch of speech engine

Page 19: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

19

TELEPHONY

Page 20: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

20

Telephony• "The working or use of telephones"• Remember, a lot of Android devices are

smartphones• android.telephony package in the API

Page 21: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

21

Page 22: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

22

android.telephony package• classes to work with• cellular network• the phone itself• SMS, short messaging service, text

messages• TelephonyManager

Page 23: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

23

Making Calls• From you app, easiest way is via Intents

• required permission is CALL_PHONE• Intent.ACTION_DIAL, brings up phone

app with numbered entered

Page 24: CS378 - Mobile Computing Speech to Text, Text to Speech, Telephony.

24

Telephony Provider• Content Provider for data related to

phone operation• tables for– carriers, MMS, SMS