diff --git a/Audio-Transcription-Chrome/content.js b/Audio-Transcription-Chrome/content.js index 1c89a2a..3c911e4 100644 --- a/Audio-Transcription-Chrome/content.js +++ b/Audio-Transcription-Chrome/content.js @@ -59,7 +59,7 @@ function init_element() { elem_container = document.createElement('div'); elem_container.id = "transcription"; - elem_container.style.cssText = 'padding-top:16px;font-size:18px;line-height:18px;top:0px;position:absolute;width:500px;height:90px;opacity:0.9;z-index:100;background:black;border-radius:10px;color:white;'; + elem_container.style.cssText = 'padding-top:16px;font-size:18px;position: fixed; top: 50%; left: 50%; transform: translate(-50%, -50%);z-index: 9999;line-height:18px;width:500px;height:90px;opacity:0.9;z-index:100;background:black;border-radius:10px;color:white;'; for (var i = 0; i < 4; i++) { elem_text = document.createElement('span'); diff --git a/Audio-Transcription-Chrome/options.js b/Audio-Transcription-Chrome/options.js index 7e8ae5b..435d75b 100644 --- a/Audio-Transcription-Chrome/options.js +++ b/Audio-Transcription-Chrome/options.js @@ -93,14 +93,10 @@ async function startRecord(option) { const socket = new WebSocket(`ws://${option.host}:${option.port}/`); let isServerReady = false; let language = option.language; - if (language === null && !option.multilingual) { - language = 'en'; - } socket.onopen = function(e) { socket.send( JSON.stringify({ uid: uuid, - multilingual: option.multilingual, language: option.language, task: option.task, model: option.modelSize diff --git a/Audio-Transcription-Chrome/popup.html b/Audio-Transcription-Chrome/popup.html index 3fae548..8d45bcb 100644 --- a/Audio-Transcription-Chrome/popup.html +++ b/Audio-Transcription-Chrome/popup.html @@ -15,112 +15,109 @@ -
- - -
- + - - - - - - + - - - - - - - - - - - - - - - + + - - - - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
@@ -134,11 +131,15 @@
diff --git a/Audio-Transcription-Chrome/popup.js b/Audio-Transcription-Chrome/popup.js index 9a708d1..e71c62d 100644 --- a/Audio-Transcription-Chrome/popup.js +++ b/Audio-Transcription-Chrome/popup.js @@ -4,7 +4,6 @@ document.addEventListener("DOMContentLoaded", function () { const stopButton = document.getElementById("stopCapture"); const useServerCheckbox = document.getElementById("useServerCheckbox"); - const useMultilingualCheckbox = document.getElementById('useMultilingualCheckbox'); const languageDropdown = document.getElementById('languageDropdown'); const taskDropdown = document.getElementById('taskDropdown'); const modelSizeDropdown = document.getElementById('modelSizeDropdown'); @@ -32,14 +31,6 @@ document.addEventListener("DOMContentLoaded", function () { } }); - chrome.storage.local.get("useMultilingualModelState", ({ useMultilingualModelState }) => { - if (useMultilingualModelState !== undefined) { - useMultilingualCheckbox.checked = useMultilingualModelState; - languageDropdown.disabled = !useMultilingualModelState; - taskDropdown.disabled = !useMultilingualModelState; - } - }); - chrome.storage.local.get("selectedLanguage", ({ selectedLanguage: storedLanguage }) => { if (storedLanguage !== undefined) { languageDropdown.value = storedLanguage; @@ -86,7 +77,6 @@ document.addEventListener("DOMContentLoaded", function () { tabId: currentTab.id, host: host, port: port, - useMultilingual: useMultilingualCheckbox.checked, language: selectedLanguage, task: selectedTask, modelSize: selectedModelSize @@ -129,7 +119,6 @@ document.addEventListener("DOMContentLoaded", function () { startButton.disabled = isCapturing; stopButton.disabled = !isCapturing; useServerCheckbox.disabled = isCapturing; - useMultilingualCheckbox.disabled = isCapturing; modelSizeDropdown.disabled = isCapturing; startButton.classList.toggle("disabled", isCapturing); @@ -142,18 +131,6 @@ document.addEventListener("DOMContentLoaded", function () { chrome.storage.local.set({ useServerState }); }); - useMultilingualCheckbox.addEventListener('change', function() { - const useMultilingualModelState = useMultilingualCheckbox.checked; - if (useMultilingualModelState) { - languageDropdown.disabled = false; - taskDropdown.disabled = false; - } else { - languageDropdown.disabled = true; - taskDropdown.disabled = true; - } - chrome.storage.local.set({ useMultilingualModelState }); - }); - languageDropdown.addEventListener('change', function() { if (languageDropdown.value === "") { selectedLanguage = null; diff --git a/whisper_live/client.py b/whisper_live/client.py index 96971c9..34fadd4 100644 --- a/whisper_live/client.py +++ b/whisper_live/client.py @@ -78,8 +78,6 @@ class Client: self.rate = 16000 self.record_seconds = 60000 self.recording = False - self.multilingual = False - self.language = None self.task = "transcribe" self.uid = str(uuid.uuid4()) self.waiting = False diff --git a/whisper_live/server.py b/whisper_live/server.py index f274608..e3498a6 100644 --- a/whisper_live/server.py +++ b/whisper_live/server.py @@ -155,7 +155,7 @@ class TranscriptionServer: options["model"] = faster_whisper_custom_model_path client = ServeClientFasterWhisper( websocket, - multilingual=options["multilingual"], + multilingual=False, language=options["language"], task=options["task"], client_uid=options["uid"], @@ -585,13 +585,11 @@ class ServeClientFasterWhisper(ServeClientBase): "tiny", "tiny.en", "base", "base.en", "small", "small.en", "medium", "medium.en", "large-v2", "large-v3", ] - - self.multilingual = multilingual if not os.path.exists(model): - self.model_size_or_path = self.get_model_size(model) + self.model_size_or_path = self.check_valid_model(model) else: self.model_size_or_path = model - self.language = language if self.multilingual else "en" + self.language = "en" if self.model_size_or_path.endswith("en") else language self.task = task self.initial_prompt = initial_prompt self.vad_parameters = vad_parameters or {"threshold": 0.5} @@ -600,7 +598,7 @@ class ServeClientFasterWhisper(ServeClientBase): if self.model_size_or_path == None: return - + self.transcriber = WhisperModel( self.model_size_or_path, device=device, @@ -620,9 +618,15 @@ class ServeClientFasterWhisper(ServeClientBase): ) ) - def get_model_size(self, model_size): + def check_valid_model(self, model_size): """ - Returns the whisper model size based on multilingual. + Check if it's a valid whisper model size. + + Args: + model_size (str): The name of the model size to check. + + Returns: + str: The model size if valid, None otherwise. """ if model_size not in self.model_sizes: self.websocket.send( @@ -635,15 +639,6 @@ class ServeClientFasterWhisper(ServeClientBase): ) ) return None - - if model_size.endswith("en") and self.multilingual: - logging.info(f"Setting multilingual to false with {model_size} which is english only model.") - self.multilingual = False - - if not model_size.endswith("en") and not self.multilingual: - logging.info(f"Setting multilingual to true with multilingual model {model_size}.") - self.multilingual = True - return model_size def speech_to_text(self):