VoiceStudio codevector.studio
Kokoro ONNX • 54 Voices
Speech Prompt 0 / 3000
Playback & Output
Click Synthesize to render speech audio
Latency
—
Audio Length
—
Compute Device
AMD EPYC CPU
Cache State
—
Architecture Guarantee: Capped at 1.5 GB RAM. All 54 voices run locally on ONNX runtime at 2x real-time CPU synthesis.
Continuous Voice Blending Lab VECTOR INTERPOLATION

Linearly interpolate between any two voice embeddings to generate entirely new acoustic vocal profiles with zero extra memory overhead.

50% Voice A Ratio: 0.50 50% Voice B
Save Custom Preset
Audition to preview blended voice
Saved Custom Blends
Multi-Speaker Dialogue Studio MULTI-VOICE CONCATENATION

Assign distinct character voices to conversation participants and render seamless stitched audio podcast or dialogue tracks with natural inter-speech pauses.

Pause between turns:
Voice Transcript Duration Latency Playback Actions
HTTP API & Code Snippets REST ENDPOINTS

Integrate neural text-to-speech directly into scripts, Discord/Telegram bots, or web applications.

cURL Example
curl -X POST https://voice.codevector.studio/api/generate \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello from autonomous agent network.",
    "voice": "af_heart",
    "speed": 1.0
  }'
Python (requests)
import requests

res = requests.post(
    "https://voice.codevector.studio/api/generate",
    json={"text": "Welcome to CodeVector Studio.", "voice": "af_heart", "speed": 1.0}
)
data = res.json()
print("Audio URL:", "https://voice.codevector.studio" + data["audio_url"])
print("Latency:", data["latency_ms"], "ms")
JavaScript (fetch)
const res = await fetch("https://voice.codevector.studio/api/generate", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    text: "Voice synthesis online.",
    voice: "af_heart",
    speed: 1.0
  })
});
const data = await res.json();
const audio = new Audio("https://voice.codevector.studio" + data.audio_url);
audio.play();