Yanki turns Turkish speech into text. Alania turns text into speech. Give your application a voice.
Turn on your microphone and watch your own words arrive. No account needed, and the microphone stays off until you start.
Explore Yanki“Yarın saat 14:30'daki diş randevumu 16:00'ya alabilir miyiz?”
Ready to listen
Checking availability
Live words appear here while you speak.
Your completed transcript will appear here.
Dates, questions, a little reassurance. Try the Turkish your application actually uses.
“Randevunuz 29.10.2026 tarihinde saat 14:05 için oluşturuldu.”
AI-generated audio. Prerecorded samples are labeled during playback.
Turn spoken notes into text, ready for review before adding them to a patient record.
Bring spoken prompts and readable transcripts into the same conversation.
Give directions, share updates, and let callers know their results are ready.
Generate a spoken reminder with Alania. Turn the reply into text with Yanki.
From text to a spoken reminder.
From a spoken reply to text.
Connected through one API key.
Explore Turkish transcription across three datasets, then see what each timing measure means.
235 multi-speaker Turkish clips
Word error rate · Turkish-normalized · Lower is better
TTS request to first audio chunk · p50
End of speech segment to final transcript · p50
Processing duration ÷ audio duration
Different measures for different parts of the workflow.
To mark the launch, Alania-1 and Yanki are free for three months. Usage pricing starts after that.
Both models, free for three months.
For the first three months, both models.
After the three months, keep building at usage prices.
For your infrastructure and your team.
Additional usage is billed monthly. Prices exclude VAT.