Majel's voice in Discovery?

[starmike]

Last year, there was a post on the Roddenberry twitter account that Majel recorded phonemes before she passed, and it was hinted that her voice would be used on Discovery. Anyone hear anything more about this?

 

[Four Lights]

Not a sausage! Between this or Marina Sirtis doing the computer voice - It's something she was lobbying for being Majel's "daughter" - I know which I'd prefer.

 

[Mr. Laser Beam]

I definitely heard that there are enough existing samples of Majel's voice that the DSC people will be able to come up with a computerized "virtual Majel" that can say any line. Not just edited from existing lines, but all new ones.

Failing that, I'm all for Sirtis doing the voice.

 

[jefferiestubes8]

I posted in 2009 specifically how it could be done using Majel's voice and that it would take some time.. on this thread:
The computer voice on next Trek TV series - actor

I definitely heard that there are enough existing samples of Majel's voice that the DSC people will be able to come up with a computerized "virtual Majel" that can say any line.

The date and source

Roddenberry.com tweeted today that Majel's voice was recorded phonetically before she passed, and that they've been working for years to get it functional for things like Siri -- and that they're currently seeing if they can get it to work for Discovery.

There are a few new software that could be used.

Adobe Reveal Project VoCo Speech Synthesis Technology

though Innoetics technology could potentially be used to make Bixby’s voice more human
http://www.techradar.com/news/samsungs-latest-acquisition-could-give-bixby-a-boost

San Francisco-based startup BabelOn believes it can bridge that gap with its speech synthesis technology

capturing a voice at the company’s studio takes about two hours and generating a range of modulation and translation capabilities could take between a week to several months.

https://thenextweb.com/tech/2017/06...te-speech-language-using-voice/#.tnw_VHn1g0PX


Lyrebird claims it can recreate any voice using just one minute of sample audio

Project VoCo requires at least 20 minutes of sample audio before it can mimic a voice, Lyrebird cuts this requirements down to just 60 seconds. The results certainly aren’t indistinguishable from human speech, but they’re impressive

https://www.theverge.com/2017/4/24/15406882/ai-voice-synthesis-copy-human-speech-lyrebird


Google’s DeepMind AI fakes some of the most realistic human voices yet

Google’s DeepMind artificial intelligence has produced what could be some of the most realistic-sounding machine speech yet. WaveNet, as the system is called, generates voices by sampling real human speech and directly modeling audio waveforms based on it,

https://www.theverge.com/2016/9/9/12860866/google-deepmind-wavenet-ai-text-to-speech-synthesis
https://deepmind.com/blog/wavenet-generative-model-raw-audio/[/QUOTE]