jm_baidu_stt_plugin 0.0.9

Baidu speech recognition & wake-up Flutter plugin powered by the latest native SDKs.

0.0.9

• iOS: Align ASR configuration with Baidu's official iOS sample (adds BDS_ASR_PRODUCT_ID and BDS_ASR_LANGUAGE, sets default strategy to online).
• iOS: Make recognition text delivery more robust (adds fallback parsing and always emits statusFinish, even when the SDK returns an empty/unknown result format).
• iOS: Reset selected per-session parameters before each start (long speech, cache audio, wakeup word) to avoid stale state.
• iOS: Improve stability for rapid start/stop (busy backoff + retry).
• iOS: Ensure PrivacyInfo.xcprivacy is included as a pod resource.

Notes:

• If you need parallel mode (online + offline), pass strategy = EVR_STRATEGY_BOTH and make sure offline model/license files are available.
• permission_handler microphone macro guidance has been moved to README.md.

0.0.8

• iOS: Fix CocoaPods resource packaging to avoid Xcode "Multiple commands produce" caused by flattening bundle resources.
• iOS: Improve iOS artifact download logic to prevent missing .a / .dat files in local dev.
• iOS: Update microphone usage description guidance.

0.0.7

• Style: Run dart format and sync versions across configuration files.

0.0.6

• Fix: Sync .podspec version with pubspec.yaml to avoid iOS install issues.

0.0.5

• Chore: Remove large binaries from Git history to keep the repository lightweight.

0.0.4

• Fix: Finalize binary distribution via GitHub Releases.

0.0.3

• Fix: Optimize package structure to meet pub.dev size limits.

0.0.2

• Feature: Add Dart APIs for init/start/stop and an event stream.
• Build: Bundle the latest native SDK libraries/resources and update Android/iOS build configuration.

0.0.1

• Initial release: Baidu Speech ASR/wake-up SDK integration (Android v3.4.5, iOS v3.0.12).