Optional cloud voices

Get an Azure Speech key for yukoe.

Create one Speech resource, copy its key and exact region, then let yukoe load the live Azure voice catalog for English, Japanese, and more.

Create a Speech resource ↗Follow the four steps ↓
F0 optionChoose the Free pricing tier when Azure offers it for your subscription and region.
Two valuesThe Speech key and its matching region identifier.
Live catalogyukoe fetches voices available to that resource.
Before you create it

The region is part of the credential.

Azure Speech keys are region-scoped. A valid key paired with the wrong region returns an authentication error, so copy both values from the same resource.

You need

  • A Microsoft account and an Azure subscription. Microsoft provides a free-account path, but account eligibility varies.
  • Permission to create a Foundry resource for Speech in the Azure portal.
  • The normal yukoe Windows app open at Settings → API & Services.
Four short steps

Create, copy both values, restart, listen.

Never paste the key into this website, Twitch chat, a screenshot, or a source file.

01

Create a Speech resource

In the Azure portal, create a Foundry resource for Speech.

  1. Choose the intended subscription and resource group, then give the resource a unique name.
  2. Choose a nearby supported region. For example, Japan East is identified as japaneast.
  3. Choose Free F0 if it is offered and you want the free hard-limited tier; otherwise stop and review pricing before choosing S0.
02

Copy Key 1 and the exact region

After deployment, choose Go to resource, then open Keys and Endpoint.

  1. Copy KEY 1 (KEY 2 also works, but use only one).
  2. Copy the region identifier shown for this resource, such as japaneast — not a translated display label and not a guessed location.
  3. Keep this page private. Regenerate a key if it appears in a stream, screenshot, log, or chat.
Check official region identifiers ↗
03

Save both values inside yukoe

In yukoe, open Settings → API & Services → Add cloud voices with provider keys.

  1. Paste KEY 1 into Azure Speech key.
  2. Paste the matching lowercase region identifier into Azure region.
  3. Choose Save securely and restart. The key goes directly to your current-user Windows DPAPI vault.
04

Select Azure and preview one voice

After restart, open Speech settings and select Azure as the engine.

  1. Choose an English and/or Japanese voice from the live catalog.
  2. Use the preview button before enabling all TTS.
  3. Turn on only the languages you want, then test one short private message. Translation and chat continue even if Azure speech is unavailable.
Read Microsoft's current TTS quickstart ↗
If it does not work

Most failures are one of these three.

401 or authentication error

The key and region usually do not match. Copy both again from the same resource; use the identifier such as japaneast, not “Japan East.”

No Azure voices appear

Confirm Azure is selected, the saved key is configured, the resource supports Speech, and network access is available. Then restart yukoe once.

Speech stops later

Check Azure Metrics, quota, and the resource pricing tier. F0 has non-adjustable limits; yukoe's chat and translation remain independent.