Our intrepid test goose, “Tank” Featherstone has been hard at work trying out some new ideas in the cockpit, and we thought that we’d jot down some of the things he’s been working on here to give you a flavor of the things that are in the pipeline for Talk to Me, Goose! We’d love to hear your thoughts about features that are needed or wanted as well.
Story Mode
OK…we’re super excited! We’ve gotten together the most talented flock of literary geese imaginable and worked up Story Mode / Story Builder / Story Time. We are, admittedly, still trying to land on a name for it. Streaming Audio unlocked and enabled this capability by allowing long form speech generation. We give every user the ability to create an unlimited number of stories, and we give every user the opportunity to convert one story into speech for free.
Yep…we’ve engaged Ernest Honkingway, William Shakesfeather, Louisa May Quackott, JD Soaringer and a host of other AI-enabled literary geese luminaries to help you craft amazing stories for a variety of different age ranges and speak them out in your own synthesized voice or in one of the many voices from the voice library…with just a few clicks and almost no typing! Here’s a video demo of the pre-release version.
We’re going to continue tweaking Merlin and the team’s story-making capabilities, but look for this in the next major version release in the coming weeks.
Pricing & Availability
We are laser-focused on working towards making this solution as accessible and available to people living with ALS, in particular, as possible. To that end, we are working tirelessly to build partnerships and relationships that will allow us to offer Talk to Me, Goose! to people living with ALS who have an ElevenLabs voice clone for free. We’ve got a plan, and we’re working the plan.
In the meantime, for those of you who are currently living with ALS, if you already have an ElevenLabs voice clone, the pricing is $4.99 / month for a “Wingman” subscription. The Wingman subscription gives you full access to all of the features of the application and your voice clone. You also have access to any other voices in the library that you may choose to use as well. For example, you just might want to build and tell a story in a pirate voice!.
The Wingman subscription cost helps to offset our costs to serve the infrastructure for the application and all of the AI-enabled features. We are actively working to find a way to make this free to use for people like you, for people like us, really. Don’t forget, this application was built by a person living with ALS and their caregiver, (by a PALS/CALS team for a PALS). We get it.
If you are PALS, and you don’t yet have an ElevenLabs voice clone, we urge you to get one. You can work with Bridging Voice and / or the Scott-Morgan Foundation to get access to ElevenLabs to get one for free.
ElevenLabs Version 3 Speech Model
In early June, ElevenLabs published an “alpha” version of its version 3 speech model, and WOW! If you watch the video demo above of Story Mode, the narration of the video was created with it. Listen carefully to it. It’s subtle, but amazing. Embedded in the playback are vocal expressions that you just can’t get with the prior models. It renders things like sighs, laughs, giggles, gasps, whispers, shouts, and the like. It allows for the embedding of narrative clues, and it interprets expressiveness from the context of the text submitted as well. We’ve been “playing” with it, and it is impressive. What would it be like to laugh again? Or giggle? Or whisper? We can’t wait to build it into the app and then see what we can do with it by building it into the full fabric of the capabilities of the app. Putting Merlin to work with the capabilities of the version 3 model unlocks some really interesting ideas. Stay tuned. If you’re curious, here’s a sample audio clip.
Building a sustainable, and hopefully fun, community
Most importantly, we continue to build a community of collaborators who are focused on making Talk to Me, Goose! available to folks who need it most. With initial meetings in the books with the leadership teams at Live Like Lou, the Scott-Morgan Foundation, and Bridging Voice, we look forward to continuing the dialog with them about how we put this application in the hands of people living with ALS globally, to reconnect people with their voice, to reconnect each of us with our loved ones and to reduce the level of effort and stress where we can. Oh, and to make it a little bit of fun too.
Android Version
We continue to hear on the Facebook page “Is there an Android version?” and the answer right now is, “No. Not yet.” (By the way, there is a Facebook page, and if you’re not following us there, you should!)
The iPad version is the most full-featured version of the app that we have incorporating all of the AI-enabled features available. The real estate on the screen allows us to provide the full predictive text capabilities, the intuitive and personalized common phrases and Merlin all on the same screen. And, the underlying Swift programming language and declarative user interface capabilities of the language makes it very dynamic to the various screen sizes and languages available in the operating system. It’s very stable as well.
Porting to the Android platform would require a fairly significant effort and a number of tradeoffs to be made, but we continue to explore this as a possibility.
Recently Released!
Message Banking
It’s out! We released Message Banking integration with Version 4 in early June. We’d love to hear your feedback on it. Let us know if you like it, love it, hate it! Leave us a review in the Apple App Store! Again, play back pre-recorded messages that are stored on your computer right in the app. You can upload them here on the website. I
Addressing Latency
We released Streaming Audio in Version 4 in early June as well. You should have noticed a reduction in the latency to render speech, especially with longer form texts right away. You should have also noticed the ability to pause, resume and stop audio playback on the main screen as well. This was a major upgrade, allowing for more natural conversation because it shouldn’t take so long to start speaking AND sometimes, when we start talking, we have to stop.
Interrupt Me
Yep…sometimes, you’ve got to be interrupted. That’s why we added “Pause / Resume” and “Stop” to the audio playback. It’s out in Version 4 as well. Version 4 was a pretty major upgrade. The cool thing about version 4 is that it set the stage for things to come.
Tell us what you think
These are just some of the things that our team is working on. We’d really love to hear from you what you want and need. Send us a note on our contact page and let us know. If these are not the things that we should be spending our time on, let us know that as well.