I have noticed that dictations shorter than 6-7 seconds get the live preview transcript but then no text is actually transcribed and the overlay says No speech detected. This is troublesome as often the sentence maybe 4-5 words or just even a word.
Another thing to note is if you start recording and then say a word or two and then don't say anything and wait for 15s, the model then transcribe and enters those words (albeit with some errors appended due to the environmental sounds captured when silent).
One should be able to dictation short sentences and even single words.
{
"api" : {
"enabled" : false,
"loopbackOnly" : true,
"port" : 8978,
"remoteAccessAllowed" : false,
"running" : false
},
"app" : {
"build" : "865",
"bundleIdentifier" : "com.typewhisper.mac",
"bundlePathKind" : "systemApplications",
"isDevelopment" : false,
"launchTimestamp" : "2026-06-16T03:29:58Z",
"uptimeSeconds" : 11326.21029496193,
"version" : "1.5.0"
},
"audio" : {
"audioDuckingEnabled" : true,
"audioDuckingLevel" : 0.1,
"defaultOutput" : {
"deviceID" : 70,
"name" : "MacBook Air Speakers",
"transportType" : "builtIn",
"uid" : "BuiltInSpeakerDevice",
"volume" : 0.43749997
},
"inputDiagnostics" : {
"defaultInputDeviceID" : 77,
"devices" : [
{
"compatibility" : "unknown",
"deviceID" : 77,
"inputChannels" : 1,
"inputOnlyCaptureFormat" : "48000.0 Hz\/1 ch\/pcmFloat32\/interleaved=false",
"isAggregate" : false,
"isAggregateOrVirtual" : false,
"isDefaultInput" : true,
"isSelected" : true,
"isVirtual" : false,
"listedByTypeWhisper" : true,
"name" : "MacBook Air Microphone",
"nominalSampleRate" : 48000,
"outputChannels" : 0,
"transportType" : 1651274862,
"transportTypeFourCC" : "bltn",
"transportTypeName" : "builtIn",
"uid" : "BuiltInMicrophoneDevice"
}
],
"previewActive" : false,
"previewAudioLevel" : 0,
"previewRawLevel" : 0,
"selectedInputDeviceID" : 77,
"selectedInputDeviceName" : "MacBook Air Microphone",
"selectedInputDeviceUID" : "BuiltInMicrophoneDevice",
"selectedInputUsesBluetoothTransport" : false
},
"mediaPauseEnabled" : true,
"selectedInputDeviceName" : "MacBook Air Microphone",
"selectedInputDeviceUID" : "BuiltInMicrophoneDevice"
},
"counts" : {
"dictionaryCorrections" : 0,
"dictionaryTerms" : 1,
"enabledProfiles" : 0,
"enabledSnippets" : 0,
"errorEntries" : 0,
"historyRecords" : 0,
"profiles" : 0,
"snippets" : 0
},
"errors" : [
],
"exportedAt" : "2026-06-16T06:38:44Z",
"model" : {
"allowsTranscriptPreviewFallback" : true,
"isModelReady" : true,
"selectedModelId" : "openai_whisper-large-v3_turbo",
"selectedProviderId" : "whisper",
"supportsLiveTranscriptionSession" : false,
"supportsStreaming" : true,
"supportsTranslation" : true
},
"permissions" : {
"accessibilityGranted" : true,
"microphoneGranted" : true
},
"plugins" : [
{
"allowsTranscriptPreviewFallback" : true,
"bundled" : false,
"currentSettingsActivity" : {
"isError" : false,
"message" : "Loading model"
},
"enabled" : true,
"id" : "com.typewhisper.whisperkit",
"isConfigured" : true,
"name" : "WhisperKit",
"providerId" : "whisper",
"registrySource" : "official",
"registryVersion" : "1.0.24",
"runtimeLoaded" : true,
"selectedModelId" : "openai_whisper-large-v3_turbo",
"source" : {
"bundleIdentifier" : "com.typewhisper.whisperkit",
"bundleShortVersion" : "1.0.24",
"bundleVersion" : "1",
"executableSHA256" : "678f49ab67a02ea8587230d2c3eaef4710b48e44348a5096eb4915af27942293",
"executableSizeBytes" : 6492176,
"pathHint" : "~\/Library\/Application Support\/TypeWhisper\/Plugins\/WhisperKitPlugin.bundle",
"sourceKind" : "managedExternal"
},
"storedLoadedModelId" : "openai_whisper-large-v3_turbo",
"storedSelectedModelId" : "openai_whisper-large-v3_turbo",
"supportsLiveTranscriptionSession" : false,
"supportsStreaming" : true,
"version" : "1.0.24"
}
],
"schemaVersion" : 6,
"settings" : {
"appFormattingEnabled" : true,
"bundledReleaseChannel" : "release-candidate",
"dockIconBehaviorWhenMenuBarHidden" : "keepVisible",
"externalStreamingDisplayCount" : 0,
"historyRetentionDays" : 0,
"indicatorStyle" : "overlay",
"indicatorSupportsTranscriptPreview" : true,
"indicatorTranscriptPreviewAvailable" : true,
"indicatorTranscriptPreviewEnabled" : true,
"indicatorTranscriptPreviewFontSizeOffset" : 0,
"memoryCaptureScope" : "allDictations",
"memoryEnabled" : false,
"modelAutoUnloadPolicy" : "afterSeconds",
"modelAutoUnloadSeconds" : 1800,
"notchIndicatorDisplay" : "activeScreen",
"notchIndicatorVisibility" : "duringActivity",
"overlayPosition" : "top",
"saveAudioWithHistory" : false,
"selectedLanguage" : "en",
"selectedUpdateChannel" : "daily",
"setupWizardCompleted" : true,
"showMenuBarIcon" : true,
"soundFeedbackEnabled" : true,
"spokenFeedbackEnabled" : false,
"translationEnabled" : false,
"watchFolderAutoStart" : false
},
"skippedExternalBundles" : [
],
"system" : {
"cpuArchitecture" : "arm64",
"localeIdentifier" : "en_IN",
"macOSVersion" : "Version 15.7.5 (Build 24G624)",
"timeZoneIdentifier" : "Asia\/Kolkata"
},
"workflows" : {
"defaultLLMProviderId" : "appleIntelligence",
"enabledCount" : 0,
"enabledWorkflows" : [
],
"totalCount" : 0
}
}
Description
I have noticed that dictations shorter than 6-7 seconds get the live preview transcript but then no text is actually transcribed and the overlay says No speech detected. This is troublesome as often the sentence maybe 4-5 words or just even a word.
Screen.Recording.2026-06-16.at.12.11.15-optimised.mp4
Steps to Reproduce
Nothing is entered at the cursor and No speech detected appears in the overlay.
Another thing to note is if you start recording and then say a word or two and then don't say anything and wait for 15s, the model then transcribe and enters those words (albeit with some errors appended due to the environmental sounds captured when silent).
Expected Behavior
One should be able to dictation short sentences and even single words.
macOS Version
macOS Sequoia 15.7.5
TypeWhisper Version
Version 1.5.0 (865) - Release Candidate
Diagnostics / Logs
{ "api" : { "enabled" : false, "loopbackOnly" : true, "port" : 8978, "remoteAccessAllowed" : false, "running" : false }, "app" : { "build" : "865", "bundleIdentifier" : "com.typewhisper.mac", "bundlePathKind" : "systemApplications", "isDevelopment" : false, "launchTimestamp" : "2026-06-16T03:29:58Z", "uptimeSeconds" : 11326.21029496193, "version" : "1.5.0" }, "audio" : { "audioDuckingEnabled" : true, "audioDuckingLevel" : 0.1, "defaultOutput" : { "deviceID" : 70, "name" : "MacBook Air Speakers", "transportType" : "builtIn", "uid" : "BuiltInSpeakerDevice", "volume" : 0.43749997 }, "inputDiagnostics" : { "defaultInputDeviceID" : 77, "devices" : [ { "compatibility" : "unknown", "deviceID" : 77, "inputChannels" : 1, "inputOnlyCaptureFormat" : "48000.0 Hz\/1 ch\/pcmFloat32\/interleaved=false", "isAggregate" : false, "isAggregateOrVirtual" : false, "isDefaultInput" : true, "isSelected" : true, "isVirtual" : false, "listedByTypeWhisper" : true, "name" : "MacBook Air Microphone", "nominalSampleRate" : 48000, "outputChannels" : 0, "transportType" : 1651274862, "transportTypeFourCC" : "bltn", "transportTypeName" : "builtIn", "uid" : "BuiltInMicrophoneDevice" } ], "previewActive" : false, "previewAudioLevel" : 0, "previewRawLevel" : 0, "selectedInputDeviceID" : 77, "selectedInputDeviceName" : "MacBook Air Microphone", "selectedInputDeviceUID" : "BuiltInMicrophoneDevice", "selectedInputUsesBluetoothTransport" : false }, "mediaPauseEnabled" : true, "selectedInputDeviceName" : "MacBook Air Microphone", "selectedInputDeviceUID" : "BuiltInMicrophoneDevice" }, "counts" : { "dictionaryCorrections" : 0, "dictionaryTerms" : 1, "enabledProfiles" : 0, "enabledSnippets" : 0, "errorEntries" : 0, "historyRecords" : 0, "profiles" : 0, "snippets" : 0 }, "errors" : [ ], "exportedAt" : "2026-06-16T06:38:44Z", "model" : { "allowsTranscriptPreviewFallback" : true, "isModelReady" : true, "selectedModelId" : "openai_whisper-large-v3_turbo", "selectedProviderId" : "whisper", "supportsLiveTranscriptionSession" : false, "supportsStreaming" : true, "supportsTranslation" : true }, "permissions" : { "accessibilityGranted" : true, "microphoneGranted" : true }, "plugins" : [ { "allowsTranscriptPreviewFallback" : true, "bundled" : false, "currentSettingsActivity" : { "isError" : false, "message" : "Loading model" }, "enabled" : true, "id" : "com.typewhisper.whisperkit", "isConfigured" : true, "name" : "WhisperKit", "providerId" : "whisper", "registrySource" : "official", "registryVersion" : "1.0.24", "runtimeLoaded" : true, "selectedModelId" : "openai_whisper-large-v3_turbo", "source" : { "bundleIdentifier" : "com.typewhisper.whisperkit", "bundleShortVersion" : "1.0.24", "bundleVersion" : "1", "executableSHA256" : "678f49ab67a02ea8587230d2c3eaef4710b48e44348a5096eb4915af27942293", "executableSizeBytes" : 6492176, "pathHint" : "~\/Library\/Application Support\/TypeWhisper\/Plugins\/WhisperKitPlugin.bundle", "sourceKind" : "managedExternal" }, "storedLoadedModelId" : "openai_whisper-large-v3_turbo", "storedSelectedModelId" : "openai_whisper-large-v3_turbo", "supportsLiveTranscriptionSession" : false, "supportsStreaming" : true, "version" : "1.0.24" } ], "schemaVersion" : 6, "settings" : { "appFormattingEnabled" : true, "bundledReleaseChannel" : "release-candidate", "dockIconBehaviorWhenMenuBarHidden" : "keepVisible", "externalStreamingDisplayCount" : 0, "historyRetentionDays" : 0, "indicatorStyle" : "overlay", "indicatorSupportsTranscriptPreview" : true, "indicatorTranscriptPreviewAvailable" : true, "indicatorTranscriptPreviewEnabled" : true, "indicatorTranscriptPreviewFontSizeOffset" : 0, "memoryCaptureScope" : "allDictations", "memoryEnabled" : false, "modelAutoUnloadPolicy" : "afterSeconds", "modelAutoUnloadSeconds" : 1800, "notchIndicatorDisplay" : "activeScreen", "notchIndicatorVisibility" : "duringActivity", "overlayPosition" : "top", "saveAudioWithHistory" : false, "selectedLanguage" : "en", "selectedUpdateChannel" : "daily", "setupWizardCompleted" : true, "showMenuBarIcon" : true, "soundFeedbackEnabled" : true, "spokenFeedbackEnabled" : false, "translationEnabled" : false, "watchFolderAutoStart" : false }, "skippedExternalBundles" : [ ], "system" : { "cpuArchitecture" : "arm64", "localeIdentifier" : "en_IN", "macOSVersion" : "Version 15.7.5 (Build 24G624)", "timeZoneIdentifier" : "Asia\/Kolkata" }, "workflows" : { "defaultLLMProviderId" : "appleIntelligence", "enabledCount" : 0, "enabledWorkflows" : [ ], "totalCount" : 0 } }