New Phone Tech Transcribes Voice Calls

Google's Pixel phone now lets users take part in a voice call without speaking. It's primarily an accessibility feature though could have some practical uses in other situations.

The feature, is an extension of the existing Live Caption feature available on many Android handsets. Until now it's mainly worked on online videos that don't have their own captioning.

It uses smart speech recognition to produce captions in close to real-time and is aimed both at people with hearing impairments and those who prefer to watch videos without sound. For example, a user might want to avoid disturbing others around them (or not share what they are watching) but not want to use earphones.

Pixel handsets will now get an upgrade to Live Caption that means it works on voice calls. In theory at least that means whatever the person on the other end of the line says will appear almost immediately as text on the screen, with no audio necessary.

Conversational Challenge

How well it works remains to be seen. One big challenge is that many online videos involve people taking care to speak clearly, slowing down if necessary, and minimizing background noise where possible. Even allowing for a "telephone voice", ordinary conversations tend to be faster and may be less formally structured and focused, in a way that could make it harder for technology to figure out any ambiguity over a particular word.

The new version of the feature will also mean the user can type a response on their handset and it will be read out as automated speech to the person on the other end of the call. That will of course make it clear the feature is in use.

Auto-Response Available

As happens with many text messaging features, there will be an option to display a range of options of likely responses that the user can select and "speak" with one tap rather than have to type them out in full. These will fit the context of the conversation, for example when the person on the other end has asked a specific question that has several most likely answers. (Source: blog.google)

Google notes the feature is aimed at "people who can't or prefer not to speak on calls." As well as people with hearing or speech limitations, it could be useful for taking part in a work conversation while on the move and in a busy environment. (Source: techradar.com)

What's Your Opinion?

Is this feature genuinely useful for people without accessibility needs? Do you find automatically generated captions accurate enough to be useful? Would you be tempted to use such a feature simply because you don't want to devote all your attention to a phone call?