Text-to-speech (TTS) is a type of assistive technology that reads digital text aloud. It’s sometimes called “read aloud” technology.
text-to-speech applications are offering an innovative solution for users to interact with content by taking it out of books and computer screens and integrating it into any environment that the user finds convenient.
Most people lead a busy lifestyle and complain that they don’t have time to read, which is totally cool, because with text-to-speech, you don’t really have to carry a book or scroll infinitely through your favorite blogs or publications when you could be spending that time doing something else.
There is a lot of companies offering TTS APIs, for Arabic languages we took three of the best APIs for now:
- Google Cloud TTS API WavNet
- Google Cloud TTS API Non-WavNet
- Read speaker API
This table shows the comparison between these APIs:
|GCP TTS API WaveNet voices||GCP TTS API Non-WaveNet voices||Read speaker|
|Close to real speech||8/10||5/10||5/10|
|Handles numbers||Very Good||Very Good||Very Good|
|Handles proper names and places||Good||Good||Good|
|Handles ambiguous words||Bad||Bad||Very Bad|
|Number of Voices||3||3||2|
|Pricing||$16.00 USD / 1 million characters||$4.00 USD / 1 million characters||starting at $4/month|
Google Cloud TTS API
Google Cloud Text-to-Speech converts text into human-like speech in more than 180 voices across 30+ languages and variants. It applies groundbreaking research in speech synthesis (WaveNet) and Google’s powerful neural networks to deliver high-fidelity audio. With this easy-to-use API, you can create lifelike interactions with your users that transform customer service, device interaction, and other applications.
1 – Pricing:
Cloud Text-to-Speech is priced monthly based on the amount of characters to synthesize into audio sent to the service.
|Feature||Monthly free tier||Paid usage|
|Standard (non-WaveNet) voices||0 to 4 million characters||$4.00 USD / 1 million characters|
|WaveNet voices||0 to 1 million characters||$16.00 USD / 1 million characters|
2 – Voices and Language:
It supports more than 180 voices across 30+ languages and variants, including the Arabic language with six voices.
You can listen to the voices from here.
3 – Max size of the request and number of the requests per minute:
Content limit: 5,000 Total characters per request
Requests limit: 300 Request limit per minute, 150.000 Characters per minute
You can try it by this code.
1 – Pricing:
From individual complete subscriptions starting at $4/month to institutional licenses, ReadSpeaker TextAid is the most cost-effective solution available today. Contact us about multi-user licenses or click here for more information about ReadSpeaker TextAid for Individuals and to sign up for a free trial.
2 – Voices and Languages:
It supports about 30+ languages including the Arabic language with tow voices (Male and Female). You can list to the voice samples from here.
Note:IBM Watson Text to Speech and Microsoft Azure and alot don’t support the Arabic language.
How we can use TTS in app, natively by using the Mobile OS (Android or iOS) itself.
There might be several ways to use TTS offline, like:
– Flutter plugins like flutter_tts or sytody or others.
– Java libraries like FreeTTS or AndroidMaryTTS or others.
– IOS libraries like iphone-tts or TTSOverview-iOS or others.
but it depends on the device support and even if it is achieved Arabic is not supported.
For Arabic language TTS APIs, it is still not good enough like English language TTS APIs. That’s because it has a lot of ambiguous words like ” التقى الرئيس الأفغاني مع نظيره الإيراني بعيد قمة الأربعين ” in this example the word بعيد should be pronounced بُعَيد not بَعِيد .
I think the best voice for news in Arabic Language is Google Cloud Text-to-Speech WaveNet Type, ar-XA-Wavenet-C voice name, it’s the best for reading numbers and ambiguous words and it’s more close to the human being.
Do you know that we use all this and other AI technologies in our app? Look at what you’re reading now applied in action. Try our Almeta News app. You can download it from Google Play or Apple’s App Store.