Frederick Abila

In Speech Lab

1. Open this voice in Speech Lab (sign in first).
2. Write or paste your script. Each line can carry its own delivery, from Subtle to Max, or leave it on Auto. Where it is switched on, Expressive mode adds tags such as [laugh] or [sigh] at the cursor, and sounds a little less like the voice than standard.
3. Set the pace, check the credit cost shown beside Generate, and press Generate.
4. Play the take, download it, or send it straight into a video project.

Stuck on the words? Ask the Director to draft a script line by line; nothing is generated until you press Generate.

From your own code

Create an API key in your account, then send text (up to 8,000 characters) or up to 200 lines, each with an expression from 0 to 1. Speed runs from 0.75 to 1.25. For Expressive mode, send "mode": "expressive" and text with tags such as [laugh] when the capabilities endpoint lists it; tags are refused in standard mode.

BASE=https://api.vidmoat.com/v1/speech
curl --fail-with-body "$BASE/generations" \
  -H "Authorization: Bearer $VIDMOAT_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  --data '{
    "voiceId": "9a567c18-b09b-4089-92c4-332ee40a9548",
    "lines": [
      { "text": "Every story starts with a voice.", "expression": 0.25 },
      { "text": "This one is ready when you are.", "expression": 0.75 }
    ],
    "speed": 1
  }'
# 202 with a job id: poll GET $BASE/generations/<id>, then GET .../audio

Read the Speech API guide for polling, downloads and errors.

From an AI agent

Agents connected to Vidmoat over MCP use the agent_workspace tool. Point it at this voice with speechSource: "community", add speechMode: "expressive" for tags where available, or tell the Vidmoat editing agent to "narrate this with Frederick Abila".

{
  "action": "speech_generate",
  "speechSource": "community",
  "speechVoiceId": "9a567c18-b09b-4089-92c4-332ee40a9548",
  "speechText": "Every story starts with a voice.",
  "speechSpeed": 1
}