Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/FieldDB/AndroidSpeechRecognitionTrainer
Speech recognition training app for low resource languages which interfaces with FieldDB corpora
https://github.com/FieldDB/AndroidSpeechRecognitionTrainer
Last synced: 3 months ago
JSON representation
Speech recognition training app for low resource languages which interfaces with FieldDB corpora
- Host: GitHub
- URL: https://github.com/FieldDB/AndroidSpeechRecognitionTrainer
- Owner: FieldDB
- Created: 2015-08-31T14:56:02.000Z (about 9 years ago)
- Default Branch: master
- Last Pushed: 2018-11-18T20:42:59.000Z (almost 6 years ago)
- Last Synced: 2024-06-19T03:04:05.014Z (5 months ago)
- Language: Java
- Homepage: https://play.google.com/store/apps/details?id=com.github.opensourcefieldlinguistics.fielddb.speech.kartuli
- Size: 14 MB
- Stars: 3
- Watchers: 7
- Forks: 5
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- low-resource-languages - Android Speech Recognition Trainer - Speech recognition training app for low resource languages which interfaces with FieldDB corpora. (Software / Utilities)
README
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.](http://img.youtube.com/vi/nULRWUMUc-I/0.jpg)](https://www.youtube.com/watch?v=nULRWUMUc-I)
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.georgianTablet uses fragments side by side:
![learn_x_tablet](https://f.cloud.github.com/assets/196199/2483261/6c4e6442-b0fe-11e3-93df-e74309100571.png)Phone uses list:
![learn_x_phone_list](https://f.cloud.github.com/assets/196199/2483266/7cb070b4-b0fe-11e3-9a42-de24f7e1be3f.png)And a separate detail screen (here with speech recognition showing)
![learn_x_phone_speech_recognition](https://f.cloud.github.com/assets/196199/2483269/837d01f0-b0fe-11e3-8707-748ab9b02022.png)### Development
Copy the sample constants file
```bash
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`
```java
public class PrivateConstants {
```Use the command line (or Android Studio) to build the app:
```bash
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
```bash
$ adb pull /sdcard/kartuli-speechrecognition
$ cd testinganonymouskartulispeechrecognition1513871391108-kartuli/audio/
$ ffmpeg -y -f s16le -ar 16k -i 000000000.raw -ac 1 000000000.raw.mp3
```