A good test for custom/domain-specific harnesses is that their agentic output should get better with more capable models in the mix. Our team has been shaping the context, data, and environment(s) for ours with great deliberation.
More here, in case you’re looking to compare notes/learn more:
https://t.co/uaKLd2AaYb
You'll rarely see me talk about the tools I am using but I am so grateful that @TrainnHQ exists and does some magical work in product videos and guides 💚
With over 10 years in Customer Education & Learning, it's hard to impress me with an AI-powered tool. What is so great about Trainn?
1. Minimal clicks - compared to other tools on the market, the UX is super fluid.
2. Correcting mistakes - Generative AI suggests text and then corrects your modifications - in any language. Loving this as you always need to do some adjustments on automatically generated guides. Here, you can write whatever comes to mind and then the AI assistant will modify your text (and correct your mistakes!)
3. Formats - Choose between video / guide / manual / article - different situations, different needs;
4. Advanced timeline to adjust video, narration, add a buffer - speed, delays, you choose.
5. Automated captions - the quality is comparable (or better?) to what YouTube offers, and we know they have been around for a bit longer!
6. Blurs - they move around when you zoom in / out in your video. If you have been editing videos with confidential information, you know how precious this is.
I tried one of the other leading video-generating product guides the other day and it took me > 3 hours to do a simple 2-minute video of a process in Notion. I had to go back-and-forth in ChatGPT to translate captions as I am not native in French. Today, I made 5 videos between 2 and 5 minutes in 3 hours with Trainn.
That's enough to make me a loyal client.
I am definitely not an affiliate or in any other way connected to Trainn. But I know the difference a quality tool can make and had to share!
PromptTTS 2: Describing and Generating Voices with Text Prompt
paper page: https://t.co/OAudxR7uXM
Speech conveys more information than just text, as the same word can be uttered in various voices to convey diverse information. Compared to traditional text-to-speech (TTS) methods relying on speech prompts (reference speech) for voice variability, using text prompts (descriptions) is more user-friendly since speech prompts can be hard to find or may not exist at all. TTS approaches based on the text prompt face two challenges: 1) the one-to-many problem, where not all details about voice variability can be described in the text prompt, and 2) the limited availability of text prompt datasets, where vendors and large cost of data labeling are required to write text prompt for speech. In this work, we introduce PromptTTS 2 to address these challenges with a variation network to provide variability information of voice not captured by text prompts, and a prompt generation pipeline to utilize the large language models (LLM) to compose high quality text prompts. Specifically, the variation network predicts the representation extracted from the reference speech (which contains full information about voice) based on the text prompt representation. For the prompt generation pipeline, it generates text prompts for speech with a speech understanding model to recognize voice attributes (e.g., gender, speed) from speech and a large language model to formulate text prompt based on the recognition results. Experiments on a large-scale (44K hours) speech dataset demonstrate that compared to the previous works, PromptTTS 2 generates voices more consistent with text prompts and supports the sampling of diverse voice variability, thereby offering users more choices on voice generation. Additionally, the prompt generation pipeline produces high-quality prompts, eliminating the large labeling cost.
Yaay! Zing by Trainn is live on @ProductHunt! 🥳
We have been heads down building the world's smartest screen recorder & editor. With Zing, you can record, edit, and publish stunning product videos with voiceovers—in mins 🚀
Let us know what you think - https://t.co/1oUZV0zoho
For the past couple of months, we’ve been working hard to improve the product video creation experience at @Trainn, and our beta is ready for launch.
You can now create how-tos & product walkthroughs automagically in a few minutes!
Join the waitlist at https://t.co/6ZGPZu4XNT
If you're looking to talk about different customer education and adoption strategies, our Portfolio Co @TrainnHQ Team is the right here for you. Ping @nvivek16 or @sumana__a or #Vaishnavi who will be around :)
Guinness 🍺 Irish Coffee ☕️ The Shamrock ☘️
And that's the city of D-U-B-L-I-N
We are going to SaaStock 2022!
Our first-ever global SaaS event.
1. Wish us luck.
2. Let us know if you'll be there :)
@SaaStock#SaaS#events#saastock
ONBOARD YOUR CUSTOMERS FASTER
We asked @wingmanforsales how.
The Customer Success team onboards new customers faster with the Wingman Academy, built on @TrainnHQ
Coming soon — Wingman's journey with Trainn 💙
Big day for us! We're live on Product Hunt. 😻 ✨
We've kept under the hood, trying to build the #SaaS world's second brain in its true sense, and it's now time to meet the larger community.
We'd really appreciate your love and support! Check us out: https://t.co/zCZL6IjanA
Increasing adoption …. Does not happen in a snap! 🪄 🦄
Life's hard! And a trusted training program 📚 to level up your customers' skills can make it all worth it! 🚀
#customertraining#customereducation#saas
https://t.co/r6CejanrBK