A new set of voices has been added to Google's Text-to-Speech and Speech Services

Google has announced upgrading it's Text-to-Speech (TTS) and the Speech Services in the Google speech engine with clearer and more natural voices.

The upgraded voice models have been rolled out to all the 421 voices in the 67 languages, according to the company.

The upgrade is automatic for TTS and Speech Services users on all the 64-bit Android devices over the next few weeks and they just have to download the latest update via Google Play Store to see the changes.

"We’ve seen a significant side-by-side quality increase with this change, particularly in respect to clarity and naturalness.”, read a Google post.

“If you are concerned your users have not updated this yet, you can check for the minimum version code, 210390644 on the package com.google.android.tts”, Google further said.

New default voices

Google further says that with the upgrade, the default voice in en-US will be changed to the one that is built using fresher speaker data, “which alongside our new stack, results in a drastic improvement.” Similarly, default voices in other languages have also been updated with new ones.

People who have not chosen a system voice and have carried on with the system defaults will now hear a “slightly different speaker”.

Moreover, the updated voice sample that Google provided is of the same duration (3 seconds) but the default female voice has changed and it feels slightly more relaxed and clearer in tone.