Skip to content

FieldDB/AndroidSpeechRecognitionTrainer

Repository files navigation

An Android app which lets the user build a custom visual and auditory vocabulary, useful for guided anomia treatment and self designed language lessons by heritage speakers.

Plugs into FieldDB databases to create language learning apps.

Video shows how heritage speakers can use field methods techniques and a Learn X app to practice with their friends and family.

Install You can see what this code base is for by installing this app (which was customized for Kartuli/Georgian heritage speakers). https://play.google.com/store/apps/details?id=com.github.opensourcefieldlinguistics.fielddb.lessons.georgian

Tablet uses fragments side by side: learn_x_tablet

Phone uses list: learn_x_phone_list

And a separate detail screen (here with speech recognition showing) learn_x_phone_speech_recognition

Development

Copy the sample constants file

cp app/src/main/java/com/github/fielddb/lessons/PrivateConstantsSample.java app/src/main/java/com/github/fielddb/lessons/PrivateConstants.java

Edit the class to be PrivateConstants instead of PrivateConstantsSample

public class PrivateConstants {

Use the command line (or Android Studio) to build the app:

export MODULE_DIR=/path/to/where/AndroidFieldDB/is
./gradlew installDebug

Create a local.properties file, for example:

# This file must *NOT* be checked into Version Control Systems,
# as it contains information specific to your local configuration.
#
# Location of the SDK. This is only used by Gradle.
# For customization when using a Version Control System, please read the
# header note.
#Sun Mar 05 18:21:33 EST 2017
ndk.dir=/Users/username/Library/Android/sdk/ndk-bundle
sdk.dir=/Users/username/Library/Android/sdk
keyPassword=apasswordforyourkeystore
storeFile=/Users/username/path/to/your/keystore

Decoding raw audio

$ adb pull /sdcard/kartuli-speechrecognition
$ cd testinganonymouskartulispeechrecognition1513871391108-kartuli/audio/
$ ffmpeg -y -f s16le -ar 16k  -i 000000000.raw  -ac 1 000000000.raw.mp3