Skip to content

Latest commit

 

History

History
72 lines (39 loc) · 1.6 KB

CHANGES.rst

File metadata and controls

72 lines (39 loc) · 1.6 KB

Changes

0.6.0 (2022-04-06)

  • Officially drop support for Python 3.6 (#69)
  • The required version of awscrt has been upgraded to ~=0.13.8 (#67)
  • Added support for language_model_name for both requests and responses. (#70)

0.5.2 (2022-03-15)

  • The required version of awscrt has been upgraded to 0.13.5 (#65)
  • NOTICE: This will be the last version with Python 3.6 support

0.5.1 (2021-10-18)

  • The required version of awscrt has been upgraded to 0.12.5 (#58)
  • Added official support for Python 3.10 (#58)
  • Allow partial results stabilization options (#54)

0.5.0 (2021-08-12)

  • The required version of awscrt has been upgraded to 0.11.24 (#47)
  • Added official support for Python 3.9 (#46)
  • Added confidence score to the Item class. (#45)
  • Added stabilization features with enable_partial_results_stabilization, partial_results_stability and Item.stable. (#48)

0.4.0 (2021-05-11)

  • The required version of awscrt has been upgraded to 0.11.15

0.3.0 (2021-04-22)

  • The required version of awscrt has been upgraded to 0.11.11 (#20, #22)

0.2.0 (2020-11-13)

  • The create_client helper function has been removed in favor of creating TranscribeStreamingClient directly.
  • Added support for the vocab_filter_name argument in start_stream_transcription. (#11)
  • Added support for multi-channel audio transcription and speaker labels (#13)

0.1.0 (2020-07-27)

  • Initial release of Amazon Transcribe Streaming SDK for Python.