Alania · Text to speech

Your text, in a Turkish voice.

Alania turns text into natural spoken Turkish. With Alania-2: 48 kHz audio, 11 built-in voices, style instructions, voice design and consent-based cloning, all on one endpoint.

48 kHz
Streaming audio
11 built-in voices
OpenAI-compatible

Try it API

Try it

Hear your own Turkish.

Whatever you type is generated live. Pick a voice, tell it how to speak, or describe a new one.

70/280

One of the built-in voices.

Ready

Alania-2 · 48 kHz

AI-generated audio. Prerecorded samples are labeled during playback.

Built-in voices.

The default female voice is alania2-f1 and the male voice is Öncel (alania2-oncel); alania2-m1 is there too. Chosen with voice; the same timbre on every request.

“Talebinizi ilgili ekibe aktardım, sonucu size kısa mesajla ileteceğiz.”

“Hesabınıza 12.480 lira 75 kuruş yatırıldı.”

“Endişenizi anlıyorum; birlikte sakin sakin bir çözüm bulacağız.”

“Size yardımcı olabilmem için randevu tarihini paylaşır mısınız?”

Nine more voices.

Four male and five female voices, Öncel, the default male voice, among them. Like F1 and M1, no voice actor recorded them: we designed candidates, then picked these nine in blind listening and named them. Put the id in voice, for example alania2-nisa; the same timbre on every request. With a style instruction they can drift a little further from their own timbre than F1 and M1.

All nine read the same line:

“Merhaba, ben sizin dijital asistanınızım. Size bugün nasıl yardımcı olabilirim?”

Nisaalania2-nisa
Female, gentle and playful
Öncelalania2-oncel
Male, mature and reassuring
Emrealania2-emre
Male, smooth and thoughtful
Fikrialania2-fikri
Male, clear and gentle
Sezginalania2-sezgin
Male, soft-spoken and breathy
Selinalania2-selin
Female, relaxed and friendly
Linaalania2-lina
Female, warm and calm
Melisaalania2-melisa
Female, bright and energetic
Nilalania2-nil
Female, warm and enthusiastic

Preview

You decide how it is said.

Write a short instruction in instructions; a few words such as “Whisper.” work best. Turkish instructions are translated to English before they reach the model. The voice stays; the delivery changes. This is in preview; it does not always follow instructions such as pace or whispering closely.

“Randevunuz yarın saat on buçukta, lütfen on dakika önce gelin.”

instructions: speaking slowly and calmly

“Bu konuda size yardımcı olmaktan memnuniyet duyarım.”

instructions: whispering softly

Preview

A new voice, no recording.

Describe its age, tone and energy in voice_description, in Turkish or English. This is in preview; the result will not always match your description exactly.

“Siparişiniz kargoya verildi, iki gün içinde elinizde olur.”

voice_description: A young man with a clear, energetic voice

Beta

Cloning, with consent.

Copy a voice from a 3–30 second recording and its transcript. Every request carries consent: true; only clone voices you have the right to use. Every response is marked as AI-generated with the x-disclosure header.

{
  "model": "alania-v2",
  "input": "Randevunuz yarın saat on buçukta.",
  "reference_audio": "data:audio/wav;base64,UklGR…",
  "reference_text": "Kaydın metni.",
  "consent": true
}

Dedicated voice (enterprise)Paid

For a closer, steadier voice, we train one just for you from 5–30 minutes of clean recordings. The speaker's consent is verified, and pricing is per voice.

Contact usor write to

API

First audio before the sentence ends.

With stream: true a WAV header arrives, then 48 kHz PCM16 as it is generated. For many sentences on one connection, use the WebSocket.

model: alania-v2 (default) or alania-v1
voice: alania2-f1 (default), alania2-oncel (male) or another built-in voice, such as alania2-nisa
instructions: How to speak, up to 200 characters
voice_description: A new voice described in words; replaces voice
stream: When true, audio arrives as it is generated (48 kHz PCM16 WAV)

API reference

curl -N https://voice.patientdesk.ai/v1/audio/speech \
  -H "Authorization: Bearer pd_live_..." \
  -H "Content-Type: application/json" \
  -d '{ "model": "alania-v2", "voice": "alania2-f1", "input": "Randevunuz oluşturuldu.", "stream": true }' \
  | ffplay -autoexit -nodisp -            # 48 kHz PCM16 as it is generated

Alania-1 is still here.

Your existing integrations keep working. Send model: "alania-v1" for Alania-1.

	Alania-2	Alania-1
Sample rate	48 kHz	24 kHz
Voices	alania2-f1, alania2-oncel and 9 more	alania
Style instructions	Yes	No
Voice design	Preview	No
Voice cloning	With consent	No
Default model	Yes	No

Let your application listen, too.

Explore Duyu