Home Artists Posts Import Register

Content

Watch the vamX.1.36.1 VIDEO (NSFW)
Direct (Recommended)
Slushe
or RedGifs (manually enable audio & switch to fullscreen)

NEW USERS:

Download vamX for AI Chat 1.36.1

IF YOU ALREADY HAVE VAMX:

1) Download the new vamX.1.36.var (or Mega). This is vamX 1.36.1.

2) Overwrite the vamX.1.36.var in your AddonPackages folder with this new file.

3) Most of the new Locations that the Chat AI can switch to are Hub Pack Scenes. If you haven't used these scenes (since they were added in vamX 1.36) they must be downloaded first. The Hub Scene Pack is automatically downloaded when any missing Hub Scene is opened (even if opened by voice or AI Chat). To manually download, press the vamX Scene tab, then the left arrow till you see a Scene page that says "Hub" on the left side (for example the page with MR Dragon Fire). Press any Scene button (that you haven't yet downloaded) and you will be notified that all Hub scenes are downloading. You can monitor the download in the Virt-a-Mate Main UI -> Package Downloader (Main UI top-middle icon showing an arrow pointing down into a drawer).

4) Open the vamX Chat AI scene. Make sure the Actions drop down is set to Actions & Feelings (or for only very, very, precise text matches try the new Exact Actions item). If you downloaded the Hub Scenes try saying "Let's go to the bar."

5) To test the Seamless Animations either check the new Chat AI Action list or go to vamX Pose -> Standing -> SEAMLESS ANIM & POSE LIBRARY. Manually opening the Seamless Animation Library preloads the animations, helps you see the voice commands, and sets some actions (like Missionary Position) that can call either poses or Seamless Animations to use Seamless Animations. If you manually opened the Seamless Animation Library, make sure that Buttons Show (on the left side) is set to Voice Commands, then click the Poses / Chilling / Hanging Out section and try saying things like "Please sit down" (or from the Dominatrix section "Yes Mistress, I'll Lick Your Feet"). You can have a single phrase trigger multiple actions, like in the video where he says "Dress up as a Dominatrix AND I'll lick your feet".

KEY NEW FEATURES

See the Documentation of ALL CHAT PHRASES which can trigger AI Chat Actions.

Improved Chat AI Action detection accuracy. This means she won't trigger many random vamX actions based on what you say / what she says. You can also set the Actions drop down to Exact Actions. See the first three items in ADDITIONAL FEATURES below.

Chat AI Actions include the entire Seamless Animation Library.

Chat AI Actions include the new vamX 1.36 Scenes / Hub Pack Scenes.

Added regular (non AI Chat) vamX voice commands for the New Scenes / Hub Pack Scenes. You can say "Let's go to the ", "Come with me to the " or "Let's have sex in the " and then say the name of the Scene to open that scene. For example for Oeshii House in the Trees you can say "Let's go to the House in the Trees".

ADDITIONAL NEW FEATURES AND FIXES

Improved accuracy of actions by removing the user and character names from the actions. For example the user name "Jack" was often triggering jumping jacks. Now if the user name is "Jack" or "Joy" it won't trigger actions like jumping jacks or happiness, etc.

The available actions have been fine tuned, removing actions that didn't make sense conversationally (like Load Clothing Set 1). This will improve the correct selection of actions (including those from the new Hub Scenes and Seamless Animations).

If you prefer to have fewer action responses (AI Chat controlled pose changes, scene changes, clothes changes, etc.) switch the Actions drop down in the Chat AI to Exact Actions (by default that drop down is set to Actions & Feelings). Exact Actions is a version of Actions & Feelings which requires a stronger (or exact) text match, so you will get far fewer Actions happening in the scene, but if any do happen, they should be much more precise.

If you speak "Retry Last" (or "Edit Last") it activates that Chat AI button.

Improved the drop downs so that they auto scroll to the latest used item. This is very important for the very long voice drop down, so you can switch to the next voice in the same section without scrolling all the way from the top.

The name of the currently selected random Chat AI is now displayed when set to NSFW Random AI (so if you like it, you can remember which one was used).

Speech servers are now faster when multiple people are accessing them at the same time.

The history is managed in a much better way for long conversations. The history now contains the latest dialog, plus the original character description, even during long conversations. This should improve AI responses for long conversations.

GPT-4 has been updated to the latest model.

Fixed an issue where if the Character and Personality text wasn't specified in the correct order, or if Personality was missing, that it would break the Character (cause incorrect responses as it thought the Character name was the entire description).

Disabled standard vamX Voice Commands while the Chat AI is open. This way, if by default you use vamX Voice Recognition, you won't get duplicated commands issued when using the Chat AI (which could sometimes cause things to break, for example if two simultaneous commands to load a new Scene are received, one from the Chat AI and one from the local vamX Voice Recognition system). This also makes sure that if you have the Chat AI set to only execute commands she agrees with, they don't execute anyway, or early, due to the local Voice Recognition system overriding the Chat AI (there is a minor issue that causes Virt-a-Mate to still display text notifications for Voice Commands even when disable by the Chat AI).

There was an occasional issue when loading a new Scene (environment), while other items were loading, that resulted in a black screen and the inability to load scenes until you restarted Virt-a-Mate. We now try to make sure there isn't anything loading before loading a new Scene from a voice command.

Files

Comments

Jayjay

First of, thanks for all your hard work! Truly appreciated. I'm having an issue with ElevenLabs in combination with different languages, where when I select a different language the option to use Elevenlabs voices disappear. It seems to work with some languages but not other, as in it seems to work with German, but not with Swedish. Is it working as intended or a bug?

vamx

Currently the ElevenLabs API v1 works with French, German, Hindi, Italian, Polish, Portuguese, and Spanish. We'll upgrade it soon to work with v2 which will add support for Chinese, Korean, Dutch, Turkish, Swedish, Indonesian, Filipino, Japanese, Ukrainian, Greek, Czech, Finnish, Romanian, Danish, Bulgarian, Malay, Slovak, Croatian, Classic Arabic and Tamil.

LawdGeddon

is that multilingual v2? If so great news :D

vamx

Yes, we will be switching soon to multilingual v2 for ElevenLabs

vamx

This is done, you can now choose any language and use it with ElevenLabs, plus we are using ElevenLabs API v2 so it supports more languages.

vamx

This is done, you can now choose any language and use it with ElevenLabs, plus we are using ElevenLabs API v2 so it supports more languages.