whisper_ggml 1.8.0 copy "whisper_ggml: ^1.8.0" to clipboard
whisper_ggml: ^1.8.0 copied to clipboard

Published 30 days ago • verified publisherantonkarpenko.com• Latest: 2.6.0
[unknown platforms]
→

Metadata

OpenAI Whisper ASR (Automatic Speech Recognition) for Flutter

More...

1.8.0 #

  • Added initialPrompt parameter to TranscribeRequest and WhisperController.transcribe
  • Wired initial_prompt through to whisper_full_params.initial_prompt on Android, iOS, and macOS to bias decoding toward domain-specific vocabulary, names, and punctuation
  • Added noContext parameter (whisper_full_params.no_context, equivalent to Python whisper's condition_on_previous_text=False) on Android, iOS, and macOS to disable cross-segment text conditioning — helps against hallucinated repetition on short utterances
  • Empty / null prompt and noContext: false leave whisper.cpp defaults, so existing callers see no behaviour change
  • Removed unused flutter_riverpod dependency, which was constraining consumers to riverpod 2.x even though the package never imported it
  • Fixed example app crash on macOS when transcribing the bundled jfk.wav (temporary directory did not exist)

1.7.0 #

  • Connected diarize transcribe parameter to the underlying whisper C++ code
  • Added diarize parameter to the transcribe method

1.6.0 #

  • Fixed iOS issues
  • Added auto language support for iOS
  • Fixed example project
  • Increased NDK version in order to support Google 16 KB requirement

1.5.0 #

  • Switched main FFmpeg from heavy ffmpeg_kit_flutter_new: ^1.6.1 to lightweight ffmpeg_kit_flutter_new_min: ^2.1.0
  • Upgraded recorder dependency for example project from v5.2.1 to v6.0.0
  • Updated main code files

1.4.0 #

1.3.0 #

  • Upgraded Android bindings to work with Flutter 3.29
  • Added new FFmpeg kit dependency

1.2.0 #

  • Fixed Android v1 embedding issue by adding override for ffmpeg_kit_flutter_full_gpl
  • Upgraded dependencies

1.1.1 #

  • Cleaned up code

1.1.0 #

  • Added support for MacOS

1.0.0 #

  • Added support for Android and iOS

← Metadata

37
likes
0
points
6.26k
downloads

Publisher

verified publisherantonkarpenko.com

Weekly Downloads

Metadata

OpenAI Whisper ASR (Automatic Speech Recognition) for Flutter

Repository (GitHub)
View/report issues

Topics

#ggml #artificial-intelligence #machine-learning

License

unknown (license)

Dependencies

ffi, ffmpeg_kit_flutter_new_min, flutter, freezed_annotation, json_annotation, path_provider, plugin_platform_interface, universal_io

More

Packages that depend on whisper_ggml

Packages that implement whisper_ggml

Back