[sg_popup id=”12″ event=”onload”][/sg_popup]
The 21st century is all about tech-savvy people. We need tech to accomplish our tasks with increased efficiency, precision & from the confines of our home, if possible. This perhaps inspired Google Cloud to design this machine learning text to speech tool, ‘Cloud Text to Speech’.
This tool offers the transformation of your written word into speech with the facility to make a choice from 30 different types of voices, coupled with a language & variant selection choice. It makes use of DeepMind’s service of groundbreaking research in WaveNet & the powerful neural networks of Google to assure the delivery of the highest fidelity & voice clarity as possible. The neural networks were engineered by the Google’s speech synthesis expertise team.
The Cloud Text-to-Speech is compatible with any application or device such as phones, personal computers, Tablets, etc, the only criteria being that the device must be able to send a REST or gRPC request. Google claims you may use this service for daily life events such as when you need it for automation of your call centers, for managing the interactive responses with the aid of Internet of Things devices, or in case of a text transformation to audio.
The exclusive features of the product are enlisted below:
- The tool provides a smooth & secure access to DeepMind WaveNet voices, which in turn ensures the most natural sounding voices with clarity and fidelity.
- It aids you to customize your speech as per the SSML tags, which makes it favorable for you to add breaks, numerical, time & period format, & other instructions for appropriate pronunciation.
- You can customize your speaking rate by 4 times, either faster or slower as per you desire.
It allows you to manually adjust the pitch of your desired voice range, to the level of 20 semitones either above or below the average output. - You can now either raise the volume of the output till the level of 16 distribution board or reduce it to the level of -96 distribution board.
- There is a choice for you with the audio formats. You can either opt for mp3 format, Linear16 format, or Ogg Plus format.
The product charges are US $4.00 / 1 million characters for Standard (non-WaveNet) voices whereas, for WaveNet voices, the costs are US $16.00 / 1 million characters. Additionally, you get a free monthly tier, the word limit for which is 0 to 4 million characters for the prior service & 0 to 1 million characters for the latter service.
Image Credit: Google