Leahnaya/SceneSwitcher
1
0
Fork 0
mirror of https://github.com/WarmUpTill/SceneSwitcher.git synced 2026-09-08 17:45:57 -05:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

Add "Speech Recognition" condition
Some checks failed
debian-build / build (push) Has been cancelled
Details
Check locale / ubuntu64 (push) Has been cancelled
Details
Push to master / Check Formatting 🔍 (push) Has been cancelled
Details
Push to master / Build Project 🧱 (push) Has been cancelled
Details
Push to master / Create Release 🛫 (push) Has been cancelled
Details

Browse Source 
Allows checking for speech patterns on a given OBS audio source
This commit is contained in:
WarmUpTill
2026-08-01 21:37:09 +02:00
committed by WarmUpTill
parent ad03b92d2d
commit 5b8d829bba
9 changed files with 1493 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
data/locale/en-US.ini
View File

@@ -840,6 +840,38 @@ AdvSceneSwitcher.condition.clipboard.condition.isImage="Clipboard contains an im
AdvSceneSwitcher.condition.clipboard.condition.isURL="Clipboard contains an URL"
AdvSceneSwitcher.condition.clipboard.condition.matches="Clipboard content matches"
AdvSceneSwitcher.condition.clipboard.condition.entry="{{conditions}}{{regex}}{{urlInfo}}"
AdvSceneSwitcher.condition.speech="Speech Recognition (beta)"
AdvSceneSwitcher.condition.speech.condition.any="Any speech is detected for"
AdvSceneSwitcher.condition.speech.condition.contains="contains phrase"
AdvSceneSwitcher.condition.speech.condition.matches="matches"
AdvSceneSwitcher.condition.speech.layout.any="{{conditions}}{{source}}"
AdvSceneSwitcher.condition.speech.layout.contains="Transcript of{{source}}{{conditions}}:"
AdvSceneSwitcher.condition.speech.layout.matches="Transcript of{{source}}{{conditions}}:"
AdvSceneSwitcher.condition.speech.layout.phrase="{{phrase}}{{regex}}"
AdvSceneSwitcher.condition.speech.layout.model="Whisper model:{{modelPath}}{{help}}"
AdvSceneSwitcher.condition.speech.model.help="GGML model files can be downloaded from https://huggingface.co/ggerganov/whisper.cpp (e.g. ggml-base.bin).\nLarger models are more accurate but slower."
AdvSceneSwitcher.condition.speech.layout.buffer="Audio buffer:{{bufferDuration}}{{help}}"
AdvSceneSwitcher.condition.speech.browse="Browse..."
AdvSceneSwitcher.condition.speech.browse.title="Select Whisper model file"
AdvSceneSwitcher.condition.speech.browse.filter="GGML model files (*.bin);;All files (*)"
AdvSceneSwitcher.condition.speech.buffer.help="Longer buffer durations improve accuracy but increase latency."
AdvSceneSwitcher.condition.speech.advanced="Advanced"
AdvSceneSwitcher.condition.speech.layout.advanced.threads="Threads:{{threads}}"
AdvSceneSwitcher.condition.speech.layout.advanced.language="Language:{{language}}{{help}}"
AdvSceneSwitcher.condition.speech.advanced.language.help="Whisper language code (e.g. 'en', 'de', 'fr') or 'auto' to detect automatically."
AdvSceneSwitcher.condition.speech.layout.advanced.translate="{{translate}}{{help}}"
AdvSceneSwitcher.condition.speech.advanced.translate="Translate to English"
AdvSceneSwitcher.condition.speech.advanced.translate.help="Translate non-English speech to English before transcribing.\nUseful when the phrase or regex is written in English but the source speaks another language."
AdvSceneSwitcher.condition.speech.layout.advanced.vad="VAD energy threshold:{{vad}}{{help}}"
AdvSceneSwitcher.condition.speech.advanced.vad.help="Minimum RMS energy a buffer must have before running inference. Buffers below this level are treated as silence and skipped. Lower values are more sensitive; raise it if inference triggers on background noise."
AdvSceneSwitcher.condition.speech.layout.advanced.suppress="{{suppress}}{{help}}"
AdvSceneSwitcher.condition.speech.advanced.suppress="Suppress non-speech tokens"
AdvSceneSwitcher.condition.speech.advanced.suppress.help="Remove filler tokens such as [MUSIC] or (applause) that Whisper tends to insert when it detects non-speech sounds."
AdvSceneSwitcher.condition.speech.layout.advanced.noContext="{{noContext}}{{help}}"
AdvSceneSwitcher.condition.speech.advanced.noContext="No context"
AdvSceneSwitcher.condition.speech.advanced.noContext.help="Do not feed the previous transcription back as a prompt for the next buffer.\nPrevents repetition across buffer boundaries at the cost of slightly reduced coherence."
AdvSceneSwitcher.condition.speech.advanced.listenWhenMuted="Listen when source is muted"
AdvSceneSwitcher.condition.speech.advanced.useGpu="Use GPU"
AdvSceneSwitcher.condition.folder="Folder watch"
AdvSceneSwitcher.condition.folder.tooltip="This condition type will allow you to monitor the contents of a folder.\nNote that the monitoring will *not* recursively scan for changes in sub directories within directories of the selected folder!\nNote that if there are several changes during a short period of time, some of the changes might not emit this signal.\nHowever, the last change in the sequence of changes always will."
AdvSceneSwitcher.condition.folder.condition.any="Any change happened"
@@ -2460,6 +2492,9 @@ AdvSceneSwitcher.tempVar.streaming.serviceName.description="The name of the stre
AdvSceneSwitcher.tempVar.clipboard.text="Clipboard text"
AdvSceneSwitcher.tempVar.clipboard.text.description="The text contained in the clipboard.\nWill be empty if the clipboard does not contain text."
AdvSceneSwitcher.tempVar.speech.speech="Transcribed speech"
AdvSceneSwitcher.tempVar.speech.speech.description="The text transcribed from the last audio buffer. Only populated when the condition matched."
AdvSceneSwitcher.tempVar.file.content="File content"
AdvSceneSwitcher.tempVar.file.date="File modification date"
AdvSceneSwitcher.tempVar.file.basename="File basename"

BIN
data/res/speech/ggml-tiny-q8_0.bin Normal file
View File

Binary file not shown.