Synthesized Vocalization
I use synthesized vocalization to speak. I have NO tongue at all, as opposed to my furry friend, Harlee. In addition to the total Glossectomy, i have a laryngectomy. I had 5 weeks between when i was diagnosed with cancer and my surgery. It was a mad scramble figuring out how i would speak 'after'. It only made sense that there was technology for this - but i mean i sure as heck didn't know anything about it! I received no guidance or input on this from medical professionals. So i was out there on my own, like so many of us, wading into the deep waters of google. I hope the information here that i have gathered, will help you along your journey.
note: YES these all cost money and i could not find any funding via insurance or other grants to fund any of this. Is this making speech available to all who need it? I should flippin' say not. So ADA mandates we have to have curb cuts for wheelchairs (i'm all for them, for the record) but screw you if you can't speak.
Synthe-who?
What is synthesized vocalization? It is the term i have chosen to describe my method of speech. (I will discuss how i came to this term in one of my forthcoming Blog entries!) There are various terms out there to describe this speech or the process; voice cloning, ai speech, voice banking, speech synthesis, synthesized speech, augmentative and alternative communication (AAC), etc. And then you have a variety of speech generating devices as well...more later. BASICALLY, you are recording your natural voice and then a company is using this data to create a synthesized version of your voice to be used in a text/type to talk application/program. This allows your 'robot' voice to be derivative of your own voice, rather than purchasing another person voice or using a stock voice. I mean no shade on a stock voice, but when I found out i was losing my voice i suddenly got very sentimental about the voice i had never been all that fond of!
Voice Banking : recording your voice with a company who will then give you a 'plug-in' that is a computer generated synthesized version of our voice used by speech generating programs/devices.
Message Banking: recording specific PHRASES in your voice. These recordings are then accessible to be played back by speech generating programs/ devices.
Speech Generating Program: the software on either your computer, tablet, or phone that generates speech from your texted/typed words.
Speech Generating Device: this can be your phone, computer, tablet or a DEDICATED tablet device made by a manufacturers solely for the purpose of typing to talk. aka AAC device.
Why bother?
Dude I'm not having anyone yank MY tongue, they are just taking my larynx. I'm totally going to be talking with a TEP in like 2 weeks, 3 tops!
DUDE, i totally hope that is the case!!!
However, what i have learned from others is that there can be issues...even laryngectomee's can lose their ability to vocalize due to complications or radiation. So to me, i think its a worth while insurance policy. I have heard from some people that their voice is already compromised or muted by the time they find out that they are going to be going under the knife. I get it. Some of the companies currently offer options where they will repair/enhance your damaged speech. I also think this field is expanding and the technology to repair will only improve. So again, insurance policy.
Where do you start?
Start with figuring out which companies you want to record with. Each company has their own process, and unfortunately seem to require/recommend different equipment to record with them. Prices range, methods of recording range (like you will need different microphone types to record for different companies...not that I WOULD ever do this, but Amazon has a really GREAT return policy, just sayin'), results differ, and where and how you can use the voices differs. Right, i know! So here are the places i decided to record with and my review of each.
This is my voice of choice. Bottom line was i got the voice that sounded most like myself.
PRICE : ranges from $235usd for the basic Voice Build, to $335 for Voice Repair, up to $400 for Voice Design (this is created from OLD recording of your voice if you didn't have a chance to record prior to losing your voice check it out!)
PRO's:
-
The version of my voice they made was the truest to my actual voice. BOOM, mic drop.
-
You can LISTEN and preview your voice BEFORE you pay.
-
They have their own app available for IOS and Android that is free. I also test drove numerous phone apps and liked thiers the best in terms of operations. You find out very quickly when you are relying on texting to speak that speed is paramount, so little things like placement of the 'buttons' mean a lot.
-
I had some questions along the way and they were responsive!
-
The voice is SAPI5 (don't ask me what that really means) which means you can use it for lots of AAC applications and hardware like those made by Tobii and Smartbox. (I will have to elaborate on these systems further elsewhere!). I do use this voice with my Tobii Communicator 5 software and it is almost fully functional. Communicator can't link in my 'message banked' phrases totally.
-
You 'own' your voice - you can download it as many times as you like and don't have 'fees' to keep it up.
CON's:
-
I did end up paying more to go up to the 'Repair' level so i could get better results, even though my voice was fairly intact.
-
There are some glitches using this voice on my Tobii Communicator 5 software
-
The message banking input is ONLY available on the phone app itself. So you have to use your phone speaker to do it (not the highest quality), and the recording to me was cumbersome. Retrieving the banked messages on the app is also rather cumbersome, i RARELY use them. REALLY RARELY.
PRICE : they gotchya by the balls buddy. $99/year for a 'subscription' or a whopping $999 perpetual license.
PRO's:
-
Not a horrible version of my voice, but a bit 'nasal' and domo aragoto mr.roboto
-
You can LISTEN and preview your voice and use your voice for a 3 month FREE TRIAL. There are real advantages to having a trial run of a voice - there are just performance issues that you really only figure out once you are knee deep.
-
The voice is SAPI and limited IOS compatible.
-
The voice is widely used and compatible with MANY phone and AAC apps.
-
Message banking of up to 300 phrases included.
CON's:
-
Domo arragato mr.roboto. They actually just sent me an improved 2023 release of my voice, i mean that's nice...that they are enhancing (it's time for me to renew my subscription coincidentally). But its still got that digital robot edge to it.
-
Subscription to my own voice? $1000 for my voice? i mean i get it, they have to keep the lights on. The real issue here is that HEALTH INSURANCE should cover our synthesized voices (i understand other countries do have means of subsidizing this)!!!! Don't get me started...just don't.
PRICE : $299 usd
PRO's:
-
My voice was pretty good. A little bit of a robot edge to it, kind of a hollow sound.
-
SAPI format makes it usable with lots of Windows communication software like PRC, Saltillo, Smartbox Accent, Tobii Communicator, etc.
CON's:
-
They also have their own iPhone app for your Voice Keeper voice, but its a super simplistic app. Um, simplistic not in a good way. Every time you type something you have to 'backspace' to erase it before you say something else. You can imagine that its hard to carry on a conversation that way. You can buy it for other apps on Android s, but i have an iPhone.
-
Just didn't like the version of MY voice as much.
PRICE : approximately $600 usd
PRO's:
-
I actually had an online meeting with several of the people in this company, and they were super nice.
-
While their policy is you need to purchase your voice site unseen (voice unheard), they extended to me a courtesty to hear my voice first. Again, super nice of them....but......
CON's:
-
My voice sounded like i was 100 years old, or maybe speaking from the inside of that tv in the Poltergeist? I'm sorry Ceraproc people! You were super nice but my voice was just horrible...horrible...sigh not much point in listing more pros and cons than that... did i say the people were nice?
PRICE : $1499 usd
PRO's:
-
? anyone?
CON's:
-
I paid $29 for the preview and wasn't overly impressed. My voice sounded fairly robotic. So honestly i didn't take it further. I had a limit to my time and resources.
-
There is some stuff that feels a bit shifty to me that i couldn't quite sort out about this company. They promote this 'human voicebank' thing like it's a social service. That you can get on and record your voice and then other people will get to use your voice or portions of your voice - like you are donating blood. However, the company isn't just donating these voices to others, they are charging $1499 and up for synthesized voices. Listen, I'm not against companies making money...i totally cannot imagine all the time it takes to do this work. But it just sits wrong how they are spinning this.
PRICE : FREE!!! yes, that's right! I said FREE. However it did highly recommend buying a specific device (Zoom H1 recorder) to record.
PRO's:
-
did i say mention this is FREE????
-
you end up recording a bunch of your personal messages and sayings that you then have in .wav format. This is also in my 'insurance policy' category. i was able to pass these on to the AI company, SpeechMorphing that i am working with to create a more hi-tech synthesized voice with. (more details on this forthcoming in section below!)
CON's:
-
to recap this is Message Banking and NOT Voice Banking.
-
it takes some work to go through the transcriptions that they return to you and edit appropriately so you can get them to work correctly on your Speech Generating device. Unfortunately because i really prefer my Speak Unique voice i can't combine and use these with that voice in my Tobii Communicator software.
-
I am unaware if there are phone apps you can use this with.
AI and the future of synthesized vocalization.
-under construction-