flutter_davoice 0.0.13
flutter_davoice: ^0.0.13 copied to clipboard
Flutter voice AI plugin for on-device STT, neural TTS, speaker verification, audio playback, and native iOS/Android voice flows.
0.0.13 #
- Live multi-speaker speaker verification:
addSpeaker,setSpeakers,removeSpeakerandgetSpeakerIdswork before start or while speech recognition runs (applied immediately, no restart).initAlltakes an optionalspeakerIdfor the onboarding JSON's speaker. getAllVoicesandgetVoiceAbilitiesfor the loaded voice model.- Refresh the bundled Android and iOS speech libraries.
0.0.12 #
- Remove unused iOS phoneme and UCD static libraries, headers, and XCFrameworks from the published package. CocoaPods now vendors only DavoiceTTS.xcframework.
0.0.11 #
- Add speaker-verification runtime options and Android speech-start options for offline preference, end-of-utterance timing, and segmented results.
- Refresh the bundled Android and iOS speech/TTS binaries and fix iOS Simulator linking of the phoneme libraries.
- Restore the previous Flutter and Dart minimums. Apply the legacy Kotlin plugin only on builds without AGP 9 built-in Kotlin, so older Flutter/AGP 8 apps and newer AGP 9 apps can use the same package version.
- Keep the method-channel map syntax compatible with the previous Dart minimum.
0.0.10 #
- Remove explicit Android Kotlin Gradle Plugin application and use Kotlin compiler options compatible with AGP 9 built-in Kotlin.
- Require Flutter 3.44 and Dart 3.12 for this release; older Flutter apps can continue using earlier package versions.
0.0.9 #
- Update the bundled Android and iOS speech/TTS artifacts with the native language and destroy fixes.
- Release native speech/TTS engines when the Flutter engine detaches and guard against an initialization completing after destruction.
0.0.8 #
- Fix Android TTS2 initialization on Android regex engines that reject Java's
(?U)Unicode character-class flag. - Preserve the native TTS2 initialization cause in Android error messages.
0.0.7 #
- Refresh the bundled Android AAR and iOS XCFramework from the latest React Native wrapper artifacts while retaining the Android API 23 manifest and ARM64 ABI contract.
- Regenerate the Maven AAR checksum sidecars for the refreshed binary.
0.0.6 #
- Lower the Android minimum SDK to API 23 while keeping the current compile SDK.
- Use normal system-recognizer microphone input on API 23–32; retain custom PCM/AEC and inline STT SpeakerID on API 33 and newer.
- Rebuild the ARM64 TTS and TTS2 JNI payloads for API 23 and document the Android ABI contract.
- Align speech pause/resume waits and readiness results with the React Native wrapper.
- Add the read-only iOS
sttHealthSnapshotdiagnostic API. - Harden iOS STT teardown against stale callbacks.
- Move Android model initialization off the method-channel thread and preserve TTS-only to full-initialization lifecycle semantics.
- Apply licenses without constructing STT; Android uses process-global licensing.
- Resolve standalone iOS TTS models asynchronously and use bounded remote model download timeouts, matching the latest React Native bridge.
- Add
clearAudioRoutingConfigand Flutter EventChannel listener diagnostics.
0.0.1 #
- Initial private Flutter plugin for Davoice STT/TTS APIs.
- Added MethodChannel and EventChannel support for Android and iOS.
- Added native iOS and Android Davoice voice engine support.
- Added support for the standalone Flutter Davoice example app.
- Added tests for channel payloads, events, and TTS/playback helpers.
0.0.2 #
- Cosmetic fixes.
0.0.3 #
- Fix Text to Speech (TTS) speed.
0.0.4 #
- Fix Text to Speech (TTS) date and words with appostrophe.
0.0.5 #
Improve TTS/STT