v1.3.11.1

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.11.1_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.11.1_win.zip

Changelog (v1.3.11.1)

• [FEATURE] Implement faster resampling method (old method is still available as audio_tools.resampy_audio()
• [FEATURE] Add new faster-whisper settings (for example: repetition_penalty to give AI a penalty for repetition.)
• [FEATURE] Add Seamless-M4T model as multi-modal model. (Speech to Text Translation + Text Translation for now)
• [FEATURE] Implement auto building.
• [BUGFIX] EasyOCR Bugfix
• [BUGFIX] nltk punkt download on pyinstaller bundle for M2M100 model
• [TASK] Update CTranslate2
• [TASK] Disable Denoise AI by default (since depending on the audio, it might filter too much, like for music)
• [TASK] Update libraries
• [TASK] Update ignorelist

Full Changelog: v1.3.10.2...v1.3.11.1

v1.3.10.2

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.10.2_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.10.2_win.zip

Changelog (v1.3.10.2)

• [FEATURE] Manage audio playback in threads, allow stopping audio
• [FEATURE] Provide way to handle playing audio via tags.
• [FEATURE] Add osc minimum frequency and queue system
• [FEATURE] Add support for custom ignorelist
• [TASK] Add osc_type_transfer option "both_inverted"
• [TASK] Add Silero fallback download server
• [TASK] Add option to force non-ui downloads
• [TASK] Add fairseq dependency
• [BUGFIX] Fix possible encoding issue with audio device names. (hopefully)
• [BUGFIX] Do not keep button values in plugin settings
• [BUGFIX] downloader removing non-zip files.
• [BUGFIX] Safely decode yaml settings
• [BUGFIX] Add sentence splitting to M2M100 text translator

Full Changelog: v1.3.9.2...v1.3.10.2

v1.3.9.2

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.9.2_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.9.2_win.zip

Changelog (v1.3.9.2)

• [BUGFIX] correctly convert chinese audio markers to wav
• [BUGFIX] calculate phrase_time_limit based on correct sample rate
• [BUGFIX] Do not add marker prompt if try_count = -1
• [BUGFIX] send valid languages struct if no STT
• [TASK] debounce settings writes
• [TASK] correctly save vad_confidence_threshold as float instead of string

Full Changelog: v1.3.9.1...v1.3.9.2

v1.3.9.1

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.9.1_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.9.1_win.zip

Changelog (v1.3.9.1)

• [HOTFIX] missing NLTK punkt dependency in standalone build
• [FEATURE] Add EU finetune models
• [FEATURE] Add silence part deletion for recordings
• [FEATURE] Add normalization
• [FEATURE] Allow disabling STT model
• [FEATURE] Implement Voice Markers as mentioned in Sharrnah/whispering-ui#10
• [BUGFIX] resample merged audio instead of each chunk seperately to prevent audio issues
• [BUGFIX] correctly splitting OSC chunks with ...
• [BUGFIX] break out of chunked sending early for new message
• [BUGFIX] Split by sentences before word tokenization for NLLB200
• [BUGFIX] fix a possible memory leak
• [TASK] Only try downloading nltk if not already existing
• [TASK] no byte conversion for enhance_audio function (to only convert if necessary)
• [TASK] use async callback for recordings

Full Changelog: v1.3.8.3...v1.3.9.1

v1.3.8.3

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.8.3_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.8.3_win.zip

Changelog (v1.3.8.3)

• [FEATURE] Add finetuned models
• [FEATURE] Add OSC chat types. (scrolling, chunk, full)
• [FEATURE] Add option to pause speech 2 text processing
• [TASK] Add length_penalty + beam_search_patience faster-whisper options
• [BUGFIX] subprocess freeze on full buffer for stdout / err

Full Changelog: v1.3.8.1...v1.3.8.3

v1.3.8.1

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.8.1_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.8.1_win.zip

Changelog (v1.3.8.1)

• [FEATURE] Add Speech 2 Speech Plugin method
• [FEATURE] Add A.I. Denoiser
• [FEATURE] Add processmanager to better handle child processes
• [BUGFIX] Fix error on base64 debug messages
• [TASK] Add additional Plugin event type
• [TASK] Update dependencies
• [TASK] add bfloat16 options

Full Changelog: v1.3.7.2...v1.3.8.1

v1.3.7.2

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.7.2_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.7.2_win.zip

Changelog (v1.3.7.2)

• [FEATURE] add global extract + move_files functions
• [TASK] Add whisper timestamp options
• [TASK] Improve download functions
• [TASK] Update modules

Full Changelog: v1.3.7.1...v1.3.7.2

v1.3.7.1

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.7.1_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.7.1_win.zip

Changelog (v1.3.7.1)

• [TASK] Update PyAudio
• [TASK] Allow disabling Text translator
• [BUGFIX] Fix EasyOCR incompatibility with Pillow v10
• [BUGFIX] Allow precision setting for M2M100
• [BUGFIX] OSC lang prefix error when using M2M100

Full Changelog: v1.3.6.7...v1.3.7.1

v1.3.6.7

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.6.7_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.6.7_win.zip

Changelog (v1.3.6.7)

• [FEATURE] Add Speech T5 Speech2Text model
• [BUGFIX] settings yaml creation on energy detection run
• [BUGFIX][#15] Download possibly missing realtime whisper model
• [BUGFIX] language listing for single language models
• [BUGFIX] Fix M2M100 + NLLB200 translators
• [TASK] Update dependencies

Full Changelog: v1.3.6.6...v1.3.6.7

v1.3.6.6

Standalone Release File (2.6 GB):

Download Server:

• EU: https://eu2.contabostorage.com/bf1a89517e2643359087e5d8219c0c67:projects/whispering/whispering-tiger1.3.6.6_win.zip

• US: https://usc1.contabostorage.com/8fcf133c506f4e688c7ab9ad537b5c18:projects/whispering/whispering-tiger1.3.6.6_win.zip

Changelog (v1.3.6.6)

• [FEATURE] Add support for specific widgets used in plugin settings
• [FEATURE] Add support for plugin widget events
• [BUGFIX] fix error on get_plugin_setting if no init_plugin_settings was used.
• [TASK] Update plugin creation readme
• [TASK] Update some dependencies

Full Changelog: v1.3.6.5...v1.3.6.6