Arabic Dialects · Speech to Text
Syrian · سوريSyrian Arabic
Speech to Text
An STT model that covers Damascene, Aleppan, and Coastal Syrian Arabic, serving Syrian speech both within Syria and across diaspora communities in Europe, Turkey, and the Gulf.
dialect: "syrian"Syrian voice. Real accuracy.18M+
Native Syrian Arabic speakers
<13.6%
WER on Syrian colloquial speech
<300ms
First-word streaming latency
4
Syrian sub-dialects covered
Try it
Speak Syrian Arabic.
Watch it transcribe.
Record yourself speaking Syrian Arabic and watch the words appear as you talk. The demo runs Hamsa's full production STT, the same model your API calls will use.
Speech to Text · speech in → text out
No sign-upYour words will appear here.
Press record and speak Arabic or English. The words appear as you talk.
Syrian Arabic dialect overview
What is Syrian Arabic?
Syrian Arabic is one of the most widely recognized and influential Levantine dialects in the Arab world, carried by landmark Damascene television drama that reached millions. It spans Damascene, Aleppan, Coastal, and Southern varieties with distinct phonological characteristics.
Key features that distinguish Syrian Arabic sub-varieties:
- Damascene, relatively soft phonology, media-prominent.
- Aleppan, harder consonants, distinct tonal character.
- Coastal Syrian (Latakia, Tartus), unique phonological features.
- Diaspora speech, influenced by German, Turkish, Dutch in respective communities.
Native speakers
18M+
Syria
Diaspora
Europe, Turkey, Gulf
Sub-dialects
4
Damascene · Aleppan · Coastal · Southern
Hamsa WER
<13.6%
on Syrian colloquial audio · 2024
The problem
Why generic Arabic STT fails
on Syrian Arabic
A model trained on MSA achieves 40–55% word error rate on Syrian Arabic. Here is why.
The solution
How Hamsa handles
Syrian Arabic natively
Capabilities
Everything you need for
Syrian Arabic STT
Transcript examples
Real Syrian Arabic.
Accurate transcription.
Use cases
Where Syrian Arabic STT
delivers results
Accuracy benchmark
Syrian Arabic speech recognition benchmark
Tested on real Syrian Arabic speech samples from diverse environments, drama, diaspora, education. Under 13.6% WER compared to 41–55% for generic models.
Word Error Rate (WER) · lower is better
Evaluated on Syrian colloquial speech benchmarks · 2024 · Lower is better
Architecture
How it works
Developer API
Two lines to Syrian Arabic speech recognition
Drop in the Hamsa SDK, set dialect: "syrian", and start streaming transcripts. No preprocessing pipeline, no normalization to MSA.
import { Hamsa } from '@hamsa-ai/sdk';
const stream = hamsa.stt.stream({
model: 'hamsa-stt-s4',
dialect: 'syrian',
diarization: true,
});
stream.on('transcript', (t) => console.log(t.text));
// → "شو عم تعمل هلق؟" · 00:01.840Related services
More Hamsa for Syrian Arabic
Why Hamsa
Built for Arabic from the ground up
FAQ
Syrian Arabic STT ·
frequently asked
Does it cover both Damascene and Aleppan dialects?
Yes. Training data comes from Damascus, Aleppo, the Coast, and the South for comprehensive coverage.
Is it suited for Syrian TV drama transcription?
Yes. The accuracy is appropriate for producing subtitles and scripts for Damascene drama and media content.
Does it work with diaspora Syrian speech?
Yes. Training data includes Syrian speech from diaspora communities in Europe, Turkey, and the Gulf.
Does it support Arabic-English or Arabic-Turkish switching?
Yes, with full support for Arabic-English code-switching.
Can it be customized with domain-specific vocabulary?
Yes. You can add custom brand names and specialized terms to improve recognition accuracy for your domain.
Explore more



