How to Enable Doubao Speech Synthesis (Cloud Read-aloud)
WhaleLink reads aloud with your phone’s built-in voice by default — free and offline. If you want the more natural, device-consistent Doubao voices, bring your own Volcano Engine API key and enable cloud read-aloud.
The process is four steps and takes about five minutes, with a one-time identity verification. Once you have the key and have activated the model, see how to enter it in WhaleLink.
Step 1: Sign up for Volcano Engine and verify your identity
Open the Volcano Engine console, register, and complete identity verification — a general requirement for cloud services.
- Console: console.volcengine.com
- Verification: personal verification is enough (ID card + face recognition).
Step 2: Activate “Doubao Speech Synthesis Model 2.0”
This is the critical step — until the service is activated, your API key has no effect on it and calls return an error.
- Activation page: console.volcengine.com/speech/new/setting/activate
- What to look for: find “豆包语音合成模型 2.0” (Doubao Speech Synthesis Model 2.0) — the version may change, pick the latest — and click 「开通」(Activate). Activating is free; you only pay per use.
Step 3: Claim the free trial quota (optional, but recommended)
New users get about 20,000 characters for six months. Find the 「试用」/「免费试用」(trial / free trial) button on the activation or product page and claim it.
Step 4: Get an API Key
- API key page: console.volcengine.com/speech/new/setting/apikeys
- Click “Create API Key”, then copy and store it. The key is usually shown in full only once.
(Optional) Use another voice: copy a voice ID from the library
WhaleLink ships several office-oriented Doubao voices — once the service is activated they work as-is. To use another voice from the library:
- Voice library: console.volcengine.com/speech/new/voices
- Find the voice, click 「复制音色 ID」(Copy Voice ID), then paste it into “Custom voice” in WhaleLink.
Enter the key in WhaleLink
Open WhaleLink → Settings → Text read-aloud (TTS) → Cloud read-aloud (Doubao) → Configure ›, tap the API Key row, paste your key, save, and tap “Test connection”.
Back on the voice list, pick a voice and tap “Preview” to hear it. Read-aloud text goes directly to Volcano Engine — it does not pass through our servers — and the API key is stored only on your phone, encrypted.
About cost
Cloud read-aloud is billed per character by Volcano Engine; you pay Volcano Engine directly, and WhaleLink is not involved in billing.
- Free quota: about 20,000 characters for six months for new users.
- Pay-as-you-go: after the quota, billed per character; see the Volcano Engine site for current pricing.
- Built-in voices: no individual purchase needed; only “voice cloning” costs extra, and WhaleLink does not support it yet.
FAQ
- Is identity verification required? Yes — a general Volcano Engine requirement.
- Do I need to buy voices? No. Activate the service and the built-in voices work; copy a voice ID to switch.
- Is my API key safe? It is stored only on your phone (encrypted) and never uploaded to our servers; read-aloud text also goes straight to Volcano Engine.
- Can I skip cloud read-aloud? Yes. The default “System read-aloud” is free and works offline; cloud read-aloud is an optional upgrade.