Excluding words
complete
Paul Schuijt
feature to ask AI to exclude certain words.
for example the typical AI words are: unparalleled , remarkable , revolutionized, underscores, journey, meticulously, testament, embark, realm, stringent, crucial, meticulous, highlights, underline, underlines, imperative, pivotal, paramount, stringent, rigors, rudimentary, orchestrated, merely, pivotal.
Autopilot
Merged in a post:
AI word quarantine
Futch ∆
Delve, Unlock, Crucial, Moreover, Dive, Elevate. �
Gil Hildebrand
I totally get where you're coming from...
But unfortunately, AI models typically don't respond well to an instruction like "never write these words". Because they are next word predictors, they can't help it but repeat things you say.
A more effective approach is to use Subscribr's powerful channel voice feature to customize your writing style. See here: https://help.subscribr.ai/article/15-controlling-your-writing-style
For example, the words "game changer" or "revolutionary" often come from an instruction to use an "engaging" or "enthusiastic" tone.
The phrases "picture this" or "imagine this" often come from a narrative technique called "vivid descriptions" or something similar.
Refining your channel voice is the best way to improve on this. If you need any help, please contact me using the chat bubble in the app.
All that said, if you still want to trying adding a list of banned words, you can do that using the Custom Instructions in the Advanced Mode while editing your channel voice.
I am leaving this thread open to help other folks discover the recommended solution.
J
Jordache Johnson
Game changer
Revolutionary
Picture this...
Imagine this...
Autopilot
Merged in a post:
‘The real kicker'
Social Media
I there a way to reduce the amount of times this phrase is used?
Thanks!
Gil Hildebrand
marked this post as
complete
Paul Schuijt
Gil Hildebrand thank you!
Gil Hildebrand
As of today you now have full control over your channel voice, including the ability to use custom instructions!
Gil Hildebrand
This is a great list Paul. I will look into expanding my list of banned words, as well as giving you some control over this as well.
Unfortunately asking AI models to exclude certain words is only somewhat effective. This is because AI models essentially love to predict the next word, so if you tell it what NOT to do, it sometimes gets confused. In some situations it can backfire. But it's certainly worth a try.
Appreciate the feedback.