Skip to content

Caedral Voice V1 · Available

See pricingAPI docs
Four voices, one owned engine. Every sample on this page is real Caedral Voice output — native 24 kHz WAV, unprocessed, from the same model that powers the API. Press play.

caedral-voice-1 · 24 kHz WAV · 16-bit mono · $0.03 / 1k chars

The arsenal

Four voices, zero filler.

Tap a sphere to hear it. Each voice ships with three reference takes, cut from the same frozen V1 identity — what you audition here is exactly what production returns.

Caedral F1

Warm narration · Female voice

caedral-f1

Caedral F2

Bright commercial · Female voice

caedral-f2

Caedral M1

Grounded explainer · Male voice

caedral-m1

Caedral M2

Crisp dialogue · Male voice

caedral-m2

Audition deck

caedral-f1 · 0:03

Voice

Reference takes

Side by side

Compare the four.

VoiceStyleLanguageLatency
Caedral F1caedral-f1Warm narrationEnglish (V1)First WAV in seconds0:03 sample
Caedral F2caedral-f2Bright commercialEnglish (V1)First WAV in seconds0:04 sample
Caedral M1caedral-m1Grounded explainerEnglish (V1)First WAV in seconds0:04 sample
Caedral M2caedral-m2Crisp dialogueEnglish (V1)First WAV in seconds0:04 sample

Samples on this page are pre-rendered and start instantly. API synthesis time scales with input length.

How it works

Text in. Audio out.

  1. 01

    Pick a voice

    Choose from the four. Each is a frozen, versioned identity — it sounds the same today and next quarter.

  2. 02

    Send your text

    Up to 2,000 characters per POST /v1/audio/speech request, at 0.5–2.0× speed.

  3. 03

    Get your WAV

    A 24 kHz, 16-bit mono file lands in seconds — download it or pipe it straight into your product.

Caedral-developed voice engine/frontend/runtime built on an attributed open-weight Kokoro foundation (Apache-2.0).

Put the arsenal to work.

Synthesize your own lines with the same four voices on plan credits or your prepaid balance — or take the API straight to production.

See pricingRead the voice docs