Write in Emirati, Saudi, Egyptian or Fusha and hear the text in the same dialect. Pauses, stress and rhythm follow the dialect, not how the words are written.
Dialect
Voice
Each one is recorded by a native speaker and tuned on how that dialect actually flows. Click a planet to hear it.
Write a tag within the sentence and the voice performs it. An expression [laugh] in Emirati does not sound like a [laugh] in Egyptian. Each tag is trained specifically for each dialect, rather than borrowed from Fusha or English.
The same tag, six different performances. Pick the dialect and the tag follows.
The same voice on the phone line, in a video, in a document read aloud, or laid over a film.
Send the text and voice ID and receive a WAV file in the selected dialect. Each request supports up to 5,000 characters.
# python import requests API = "https://api.echonova.ai" AUTH = dict(Authorization="Bearer YOUR_API_KEY") res = requests.post( API + "/v1/text-to-speech/sync", headers=AUTH, json=dict( text="مرحبا، أنا صوت من الإمارات", voiceId=VOICE_ID, # from /v1/text-to-speech/voices speed=1.0, ), ) audio_url = res.json()["data"]["audioUrl"] # a WAV file
Security & sovereignty
Speech to TextConvert audio to text in real time or in batches, across eight dialects.
Text to SpeechNatural Arabic voices in the caller’s own dialect.
Noise and echo are removed before the model listens to the audio.
From calls, apps or recordings.
Enter Arabic text in the demo, choose an available voice and generate a sample. The voice library lets you listen before choosing.
The developer documentation includes API requests, authentication and examples for integrating speech generation.
Sovereign Arabic voice AI
