Over the weekend I experimented with Lyrebird in attempt to recreate my voice to publish a podcast entirely via text. The results weren't bad but didn't sound natural enough.
I see you're exploring this space as well. How do you plan to create custom voices, @zohaibahmed / @saqibmuhammad2?
@rrhoover We're also taking a Deep Learning approach, there is a lot of excitement in this space, and I think it's just a matter of time that speech synthesis will be as good as some of the computer vision models that are out there.
Our existing solution is to work with customers more directly to get good quality data that we can use for speaker adaptation on our existing models. It's definitely a work in progress, and we hope to make it available through a UI for everyone one day when the results are consistent. For the moment, we're taking on selected pilots.
Okay. So, this is interesting. I was searching for a specific tool online and remembered PH. I stumbled upon this and I think I can use this for my future projects. Thank you. Will try later.
Hi PH! Thanks for the hunt Robleh! 👋
I'm the creator of Voicepods, a service that converts written content to human-like audio using state-of-the-art AI models.
I started Voicepods to allow anyone to be able to convert any written work into a podcast - but it's turning out to be a lot more than that. I've seen Voicepods be used for creating audio stories, product voice-overs, as well as of-course converting blogs to podcasts.
Here's a list of some cool things you can do:
1. Create realistic audio from text in 8 voices (male, female, and different accents)
2. Create a podcast that will be hosted by Voicepods. Everytime you convert another piece of text, it'll push it to your favorite podcast app like iTunes, Google Play Music, Pocket Casts, etc.
3. Publish to Google Assistant with One-Click. It'll automatically be published under the Voicepods Action, and it takes 3 days for us to get it approved as a custom action.
4. Add as a widget on Medium, Wordpress, your own website, or practically anywhere else on the web!
We're also actively doing pilots with custom voice models - so if that's your thing, give me a shout!
I would love to hear your feedback. 😁
Report
LOVE this. The stock voices are fairly good, I know you are trying to improve them. I have the same Qs as @rrhover, wondering if/when you will be doing custom voicing/personal voice printing. As I was messing around with the product, I noticed a few things: After signup, and in the projects section app.voicepods.com/projects, there is no longer a way to test the voices without "recording" one; also, the pricing tiers and features are no longer accessible when signed in. Might be worth linking those into the profile area to aid in conversion. Cool product, looking forward to testing it out more.
@rrhover@michaelstjames Thanks for the feedback Michael. Glad you like what we have so far! We are doing pilots for custom voices at the moment. I'll send you an email so we can kickoff a talk about what you're looking for.
Thanks for the other feedback! I'll be cleaning that up so we can make it easier for users to use the product.
Report
Really cool! Would be more inviting if there was a way to test it with custom input(you could limit the characters) on the landing page instead of having to sign up.
@harjapuppal appreciate the feedback :). Yes, we will be implementing that feature in the near future so users can test out the different voices.
Report
Awesome product!
I always wanted a tool that can convert images into text, something that I can take pictures with, then the tool recognise sections with texts in the picture so I don't have to take notes manually, instead, I could copy and paste everything in my Evernote in seconds.
I don't know if you have any plans to build a tool like that, but it would be so helpful!
Product Hunt
Resemble AI
VoicePods
Resemble AI
Resemble AI
Resemble AI
VoicePods