Arabic Dialects · Text to Speech
MSA · فصحىText to Speech
in Modern Standard Arabic
Broadcast-standard MSA voices, formal, educational, and news-quality, for Arab media, e-learning, government services, and official applications across all Arab countries.
dialect: "msa"4 voices · streaming · SSML · diacritics4.1 / 5
MOS naturalness on MSA speech
4
Broadcast and educational voices
<300ms
First audio chunk latency
MENA
Region-wide official language coverage
Try it
Type Modern Standard Arabic.
Hear it in a broadcast voice.
Enter any MSA text, with or without diacritics, and hear it in a natural voice. The demo uses Hamsa's MSA voice, the same model available through the API.
Text to Speech · text in → speech out
No sign-upType anything, pick a voice, and press play.
Register overview
Synthesizing Modern Standard Arabic
Modern Standard Arabic is the formal, learned register of Arab media, education, courts and government across 22+ countries. It is nobody's mother tongue, but audiences everywhere in the Arab world understand it. For news, e-learning and official services, the voice has to meet broadcast standards: every letter articulated, every vowel where it belongs.
Key features Hamsa TTS reproduces natively:
- Precise articulation of every letter, including the emphatic consonants and the qaf as /q/.
- Case endings voiced in formal reading, and word endings dropped at a pause.
- Numbers and dates read with the gender and case agreement MSA requires.
- Diacritics that settle ambiguous words: "عِلْم" (knowledge) versus "عَلَم" (flag).
Reach
420M+
Across the MENA region
Hamsa MOS score
4.1/5
naturalness on MSA speech
Streaming latency
<300ms
first audio chunk
Available MSA voices
4
broadcast & educational · female & male
The problem
Why generic Arabic TTS falls short
of broadcast MSA standards
In formal content, listeners catch a misplaced vowel immediately, and trust the voice less.
The solution
How Hamsa synthesizes
broadcast-standard MSA
Modern Standard Arabic voices
Four broadcast-standard MSA voices
Two broadcast voices and two educational voices. All supporting streaming, SSML, diacritics, and style control.
Capabilities
Everything you need for
Modern Standard Arabic TTS
Voice examples
Formal Arabic.
Synthesized naturally.
Use cases
Where MSA TTS
delivers results
Naturalness benchmark
A natural MSA voice at the highest quality
Evaluated by Arab listeners on a 5-point Mean Opinion Score (MOS) scale. Higher score means more natural-sounding speech.
MOS Naturalness Score · higher is better (max 5.0)
Mean Opinion Score rated by Arab listeners from multiple regional backgrounds · 2024 · Higher is better
Architecture
How it works
Developer API
Three lines to Modern Standard Arabic speech synthesis
Pass your MSA text, with or without diacritics, pick a broadcast voice, and stream audio. No manual phonetic transcription.
Read the API docs →import { Hamsa } from '@hamsa-ai/sdk';
const audio = await hamsa.tts.synthesize({
text: 'أهلًا بكم في نشرة الأخبار.',
voice: 'nasser-msa',
dialect: 'msa',
style: 'professional',
});
// → streams PCM audio · first chunk <300msRelated services
More Hamsa for Modern Standard Arabic
Why Hamsa
Built for Arabic from the ground up
FAQ
MSA TTS,
frequently asked
Why choose MSA instead of a local dialect voice?
MSA is the language of media, education, law and official contexts in every Arab country, and Arab audiences everywhere understand it. That makes it the right fit for news bulletins, educational content and government services aimed at more than one country. For consumer apps and everyday conversation, a local dialect voice may feel closer to the user.
Does Hamsa TTS support vocalized (diacritized) text?
Yes. You can pass fully vocalized text to set every case ending precisely, or add diacritics only to ambiguous words. This matters in educational and official content, where every letter must be pronounced exactly as intended.
How fast is Hamsa's Modern Standard Arabic TTS?
Hamsa streams audio in real time. The first audio chunk arrives in under 300ms from the API call, so it works for interactive services as well as content production.
Which MSA voices are available?
Four voices: Nasser and Nour in a broadcast style for news and official programmes, and Abdulrahman and Sara in an educational style for e-learning and training content. Hamsa voices score 4.1 on the Mean Opinion Score (MOS) scale for naturalness.
Does it support SSML and a custom pronunciation lexicon?
Yes. SSML lets you control pauses, delivery speed and pitch, and the pronunciation lexicon lets you set how proper names, institutions, and modern scientific and technical terms are pronounced.
Explore more



