web-speech-cognitive-services - npm Package Versions

[8.0.0] - 2024-11-26

Changed

• 💥 Modernized some code with TypeScript, more type-aligned to W3C Speech API, and moved to official Event Target API, in PR #220, #224, #225, and #228
  • SpeechRecognitionResult and SpeechRecognitionResultList is now a array-like object, use Array.from() to convert them into an array
• Updated build tools and added named exports via CJS/ESM
• Bumped dependencies, in PR #216 and #218
  • Production dependencies
    • base64-arraybuffer@1.0.2
    • event-as-promise@2.0.0
    • event-target-shim@6.0.2
    • memoize-one@6.0.0
    • on-error-resume-next@2.0.1
    • simple-update-in@2.2.0
  • Development dependencies
    • @types/node@20.13.0
    • @typescript-eslint/eslint-plugin@7.11.0
    • @typescript-eslint/parser@7.11.0
    • classnames@2.5.1
    • dotenv@16.4.5
    • eslint-plugin-react@7.34.2
    • jest-environment-jsdom@29.7.0
    • microsoft-cognitiveservices-speech-sdk@1.36.0
    • node-fetch@2.7.0
    • on-error-resume-next@2.0.1
    • prettier@3.3.0
    • react-dom@18.3.1
    • react-redux@9.1.2
    • react@18.3.1
    • redux-saga@1.3.0
    • redux@5.0.1
    • typescript@5.4.5

Fixed

• Fixed #218. Speech recognition should stopping properly in some cases, in PR #218
  • Interactive mode, muted microphone
  • Continuous and interactive mode, stop shortly after start
• Fixed #221. Continuous mode with successful interims should stop without errors, in PR #222
• Fixed #226. createSpeechServicesPonyfill should return both SpeechRecognition and SpeechSynthesis ponyfill, in PR #227
  • 💥 createSpeechServicesPonyfill will throw if the browser does not support Media Capture and Streams API, instead of warning and continue