Cloud TTS for VSCode
Right-click selected text (in an editor or in the integrated terminal) → "Read Aloud" → audio plays via afplay on macOS.
Supports Gemini, OpenAI, and ElevenLabs as TTS backends — pick whichever you like the sound of best.
Why
Native macOS say voices are robotic. Cloud providers sound human:
- Gemini 2.5 — natural prosody, supports natural-language style prompts ("Say sarcastically:")
gemini-2.5-flash-preview-tts (default) — faster, cheaper. ~$10/1M charsgemini-2.5-pro-preview-tts — better prosody. ~$80/1M chars (≈8× flash)
- OpenAI — fast, cheap, decent quality
gpt-4o-mini-tts (default) — newer, supports instructions. ~$0.015/min audiotts-1 — cheap and fast, no instructions. $15/1M charstts-1-hd — higher-quality variant of tts-1. $30/1M chars
- ElevenLabs — top-tier voice cloning quality (most expensive)
eleven_multilingual_v2 (default) — highest quality multilingual. ~$120/1M chars (API)eleven_turbo_v2_5 — lower latency, slightly less expressive. ~$60/1M chars (API)eleven_flash_v2_5 — lowest latency, cheapest. ~$60/1M chars (API)
Prices are approximate ballparks for rough comparison — see each provider's pricing page for exact rates and any free-tier quotas.
Install
cd ~/Documents/vsc-extensions/cloud-tts
npx --yes @vscode/vsce package --allow-missing-repository
code --install-extension cloud-tts-*.vsix # stable VSCode
code-insiders --install-extension cloud-tts-*.vsix # Insiders (if you use it)
Re-run after edits to update. Reload windows with Cmd+Shift+P → Developer: Reload Window.
First-time setup
- Set an API key:
Cmd+Shift+P → Cloud TTS: Set API Key… → pick provider → paste key. The input field is masked, and the key is stored in VSCode's encrypted SecretStorage — never written to settings.json, never synced.
- Pick a provider (if not Gemini):
Cmd+Shift+P → Cloud TTS: Switch Provider, or set cloudTts.provider in Settings.
- (Optional) tune voice/model:
Cmd+Shift+P → Cloud TTS: Open Settings.
Usage
| Action |
How |
| Read selection (editor) |
Right-click → Read Aloud · Cmd+K Cmd+R |
| Read selection (terminal) |
Right-click → Read Aloud · Cmd+K Cmd+R |
| Stop playback |
Cmd+K Cmd+S · Cloud TTS: Stop Playback |
| Quick switch provider |
Cloud TTS: Switch Provider |
| Set / change API key |
Cloud TTS: Set API Key… (masked input) |
| Delete all stored keys |
Cloud TTS: Clear All API Keys |
| Open settings |
Cloud TTS: Open Settings |
| Cancel during synthesis |
Cancel button on the progress notification |
Terminal selection is read via a brief clipboard round-trip (the only way without an official API). The clipboard is restored to its previous content immediately after.
Settings reference
Top-level
| Setting |
Default |
Notes |
cloudTts.provider |
gemini |
Active provider |
API keys are not in Settings — use Cloud TTS: Set API Key… instead (encrypted).
Gemini
| Setting |
Default |
Notes |
cloudTts.gemini.model |
gemini-2.5-flash-preview-tts |
Or gemini-2.5-pro-preview-tts for better prosody |
cloudTts.gemini.voice |
Kore |
30 voices: Zephyr, Puck, Charon, Kore, Fenrir, Leda, Orus, Aoede, … |
cloudTts.gemini.stylePrompt |
(empty) |
Prepended to text. e.g. "Say in a calm, neutral tone:" |
OpenAI
| Setting |
Default |
Notes |
cloudTts.openai.model |
gpt-4o-mini-tts |
Or tts-1, tts-1-hd |
cloudTts.openai.voice |
nova |
alloy, ash, ballad, coral, echo, fable, onyx, nova, sage, shimmer, verse |
cloudTts.openai.instructions |
(empty) |
gpt-4o-mini-tts only. e.g. "Speak cheerfully." |
ash / ballad / coral / sage / verse only work with gpt-4o-mini-tts.
ElevenLabs
| Setting |
Default |
Notes |
cloudTts.elevenlabs.model |
eleven_multilingual_v2 |
Or eleven_turbo_v2_5, eleven_flash_v2_5 |
cloudTts.elevenlabs.voiceId |
21m00Tcm4TlvDq8ikWAM (Rachel) |
Browse https://elevenlabs.io/app/voice-library for IDs |
Limitations
- macOS only (uses
afplay).
- No streaming — waits for full clip before playback. Long selections take a few seconds before audio starts.
- Terminal selection requires a clipboard round-trip (clipboard is restored after).
| |
