Package com.keenresearch.keenasr

package com.keenresearch.keenasr

The KeenASR SDK for Android is a complete set of tools, libraries, and documentation that helps developers perform speech recognition on Android devices.

KeenASR for Android currently runs on Android 15+ (or higher).

Typically, you will follow these steps when using the SDK:

1. Install ASR Bundle from the APK using KASRBundle class or download it from your own website.
2. Initialize the SDK with initWithASRBundleAtPath method.
3. Configure KASRRecognizer options and callbacks. See onPartialResult and onFinalResponse callbacks.
4. Create one or more decoding graphs. See KASRDecodingGraph class for relevant methods.
5. Prepare recognizer for listening with a specific graph. See Initialization and Prepare category for different prepareForListening* methods.
6. Call startListening to start audio capture and recognition. The onPartialResult method will be called periodically and you can use it to display what has been recognized so far. The onFinalResponse method will be called when the recognizer stops listening, and it will provide the final response (KASRResponse), which will include the final result (KASRResult).

Installation

Visit the SDK Documentation's KeenASR SDK for Android Installation page.

Quick Start

Visit the SDK Documentation's KeenASR SDK for Android Quick Start page.

Class	Description
KASRAudioQualityResult	KASRAudioQualityResult class contains various metrics for audio quality estimation, returned as part of the KASRResponse, including Signal to Noise Ratio (SNR) and various signal level metrics.
KASRBundle	An instance of the KASRBundle class, called asrBundle, manages ASR Bundle resources on the device file system
KASRDecodingGraph	KASRDecodingGraph class manages decoding graphs in the filesystem.
KASRDecodingGraph.KASRSpeakingTask	KASRSpeakingTask defines a type of interaction which might be useful when building a decoding graph, so that the way decoding graph is built can be fine tuned to the task.
KASRPhone	KASRPhone represents a phone (speech segment that possesses distinct physical or perceptual properties and serves as the basic unit of phonetic speech analysis).
KASRRecognizer	An instance of the KASRRecognizer class, called recognizer, manages recognizer resources and provides speech recognition capabilities to your application.
KASRRecognizer.KASRRecognizerLogLevel	These constants indicate the log levels for the framework.
KASRRecognizer.KASRRecognizerState	These constants indicate different states recognizer can assume.
KASRRecognizer.KASRVadParameter	These constants correspond to different Voice Activity Detection parameters that are used for endpointing during recognition.
KASRRecognizerListener	KASRRecognizerListener provides an interface that your class needs to implement in order to receive partial and final result callbacks from the KASRRecognizer instance.
KASRRecognizerTriggerPhraseListener	KASRRecognizerTriggerPhraseListener provides an interface that your class needs to implement in order to receive trigger phrase callbacks from the KASRRecognizer instance.
KASRResponse	KASRResponse contains various metadata related to the single interaction with the speech recognition system, from calling startListening until the recognizer stopped listening.
KASRResult	An instance of the KASRResult class, called recognition result, provides results of the recognition.
KASRUploader	KASRUploader class provides methods that manage uploads of speech recognition metadata and audio recordings to Dashboard, a Keen Research cloud service for data collection and development of voice applications.
KASRWord	An instance of the KIOSWord class, called word, provides word text, timing information, and the confidence of the word.
KASRWordPronunciation	KASRWordPronunciation is a class that defines a mapping between a word and its phonetic pronunciation.