日本語

SpeechDock — Advanced Features

This page covers features that require API keys from cloud providers. These are optional enhancements — SpeechDock works fully with macOS native STT/TTS without any API keys.

API Key Setup

To use cloud providers, configure API keys in Settings > API Keys:

API keys are securely stored in macOS Keychain. Alternatively, you can set environment variables for development.

Cloud STT Providers

Cloud providers offer higher accuracy, more language support, and specialized features compared to macOS native STT.

Select the provider in Settings > Speech-to-Text.

Cloud TTS Providers

Cloud TTS provides natural-sounding voices with various styles and languages.

Voice and Model Selection

Each provider offers different voices and models. Select them in:

• Settings > Text-to-Speech (persistent setting)
• TTS Panel header (quick switch)

Audio Output Device

Route TTS playback to any audio output device (speakers, headphones, virtual devices). Select in Settings > Text-to-Speech or the TTS panel.

Audio File Transcription

Transcribe pre-recorded audio files. Available with cloud STT providers and macOS native (macOS 26+). Not available with Grok provider.

Note: macOS native file transcription requires macOS 26 or later and processes audio entirely on-device — no API key or internet connection needed.

How to Transcribe

Drag & Drop: Drag an audio file onto the STT panel’s text area.

Menu Bar: Select Transcribe Audio File… from the SpeechDock menu bar.

The STT panel placeholder displays the supported formats and limits for the currently selected provider.

Translation with External Providers

While macOS on-device translation supports ~18 languages, cloud providers offer:

• 25+ languages (all languages in the language list)
• Higher translation quality using LLMs
• Works on macOS 14+ (no macOS 26 requirement)

Translation Providers and Models

Switching Translation Provider

• Settings > Translation: Set the default provider and model
• Panel: Click the ⚡ button next to the translation controls for quick switching

Provider Auto-Sync

When you switch STT or TTS providers, the translation provider automatically syncs:

Subtitle Real-time Translation

When using subtitle mode, you can enable real-time translation that translates speech as you speak. This works with all audio sources (microphone, system audio, app audio).

How It Works

1. Enable subtitle mode (Ctrl + Option + S)
2. Click the globe icon (🌐) in the subtitle header to enable translation
3. Select your target language and translation provider
4. Start recording — translations appear in real-time

Translation Providers for Subtitles

Note: Subtitle translation uses the provider’s default model for optimal performance. This is independent of the model selected in the panel translation settings.

Features

• Caching — Repeated phrases are translated instantly from cache (up to 200 entries)
• Context-aware — LLM providers use recent sentences as context for better translations
• Pause detection — Automatically triggers translation after 1.5 seconds of silence
• Settings sync — Translation settings sync from the STT panel when subtitle mode starts

Limitations

• Translation adds some latency compared to transcription-only mode
• Cloud providers require API keys and internet connection
• macOS provider requires macOS 26+ and downloaded language packs

Language Selection

Both STT and TTS support language selection with all cloud providers:

• Auto (default): Automatically detects the spoken/target language
• Manual: Choose from 25+ supported languages

Available languages: English, Japanese, Chinese, Korean, Spanish, French, German, Italian, Portuguese, Russian, Arabic, Hindi, Dutch, Polish, Turkish, Indonesian, Vietnamese, Thai, Bengali, Gujarati, Kannada, Malayalam, Marathi, Tamil, Telugu.

TTS Speed Control (Save Audio)

When saving audio to a file, speed is controlled differently from real-time playback:

For real-time playback, speed is always controlled locally via audio processing, allowing dynamic adjustment during playback.

Privacy Considerations

When using cloud providers:

• Audio data is sent to the respective provider’s API for processing
• Each provider has its own privacy policy and data retention rules
• For maximum privacy, use macOS native providers (all processing on-device)
• API keys are stored in macOS Keychain and never shared between providers

---