https://livekit.io logo
Join Slack
Powered by
what is the latest stable version of livekit in js...
# ask-ai
b
what is the latest stable version of livekit in js for production environment
t
Thanks for reaching out, Harsh. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.
Just to keep you in the loop, the latest stable version of LiveKit for JavaScript that’s great for production use is 2.15.6. If you need any help with it, I’m here for you! Sources: CHANGELOG.md
b
how to install it
t
I'd be happy to help you get started with the latest stable version of the LiveKit JavaScript SDK! You can choose one of the following easy methods to install it: • Using npm: 
npm install livekit-client --save
• Using yarn: 
yarn add livekit-client
• Using a CDN: 
<script src="https://cdn.jsdelivr.net/npm/livekit-client/dist/livekit-client.umd.min.js"></script>
This will make the module available under the global 
LivekitClient
namespace. If you have any questions or need a hand with anything else, just let me know—I'm here to help! Sources: Connecting to LiveKit | LiveKit Docs | Docs::Home | Docs::Home | JavaScript quickstart (web) | LiveKit Docs | package.json
b
what is the latest version for livekit agent
t
Thanks for reaching out! The latest version of LiveKit Agent actually depends on the platform you're using: • Node.js/TypeScript: The latest version is 1.0.2. • Python: The latest version is 1.2.8. Whenever you get a chance, could you please let me know which platform you're working with? That way, I can make sure to give you the most helpful and accurate information. Looking forward to helping you out! Sources: plugins/test/CHANGELOG.md | livekit-agents/livekit/agents/version.py | Agents Overview | LiveKit Docs
b
why is your code base doesnt have modalities text option
t
I'm happy to share that LiveKit does support the 
text
modality option in its codebase. You can set the 
modalities
parameter to include 
text
either on its own or alongside 
audio
, depending on what output mode you'd like to achieve. • If you set 
modalities=["text"]
, it will enable text-only mode. • By default, 
modalities
is 
["text", "audio"]
, but you can certainly configure it explicitly for text-only use if that suits your needs better. This feature is also nicely demonstrated in example implementations like the Realtime Playground and Gemini Playground. Please feel free to reach out if you'd like any help with setting this up—I’m here to assist! Sources: livekit-plugins/livekit-plugins-openai/livekit/plugins/openai/realtime/real... | agent/main.py | web/src/components/code-viewer.tsx | OpenAI Realtime API integration guide | LiveKit Docs | OpenAI Realtime API integration guide | LiveKit Docs
b
why is it showing that this is not available
t
I'd love to help you out! Could you please share a bit more about the issue you're facing? Specifically: • Are you seeing any error messages? If so, it would be great if you could share the exact text. • Which version of the LiveKit SDK or platform are you currently using? • How are you setting up the 
modalities
parameter in your code? Once I have these details, I'll be able to assist you much better. Looking forward to your reply!
b
we are using 1.0.2 version
src/agent-openai.ts7839 - error TS2353: Object literal may only specify known properties, and 'modalities' does not exist in type '{ model?: string; voice?: string; temperature?: number; toolChoice?: ToolChoice; baseURL?: string; inputAudioTranscription?: InputAudioTranscription | null; ... 7 more ...; connOptions?: APIConnectOptions; }'. 783 modalities: ["text"], this is the error
t
I'm glad you're exploring the 'text' modality option in LiveKit version 1.0.2! It should be available, but sometimes a few common hiccups can get in the way. Let’s walk through them together: • Configuration Issue: Double-check that you’ve explicitly set 
modalities=["text"]
when initializing the 
RealtimeModel
. This step is key to making sure text is enabled. • Audio Output Conflict: If 
audio_enabled=True
in 
RoomOutputOptions
, the system might prioritize audio output over text. Try setting 
audio_enabled=False
to see if that helps. • Frontend Subscriptions: It’s important to confirm your frontend is subscribed to the right topics, like 
lk.chat
for text messages and 
lk.transcription
for transcriptions. This ensures everything flows smoothly. • Version-Specific Bug: Although no specific issues are known for version 1.0.2, there’s always a chance of a bug or inconsistency. Running the 
realtime_with_tts.py
example might be a great way to verify your setup. If you’ve gone through these steps and things still aren’t working as expected, please don’t hesitate to reach out. I’m here to help and we’ll get this sorted together! Sources: Realtime model integrations | LiveKit Docs | OpenAI Realtime API integration guide | LiveKit Docs | livekit-agents/livekit/agents/voice/room_io/room_io.py | Text and transcriptions | LiveKit Docs
It looks like the error is telling us that the 
modalities
property isn’t recognized in the object type you’re using. This might happen because the version of the LiveKit SDK you have (1.0.2) doesn’t support the 
modalities
property in the context of 
agent-openai.ts
. Here are a few steps that might help you resolve this: • First, double-check if the 
modalities
property is actually supported in your SDK version. If it’s not, updating to a newer version where this property is included could do the trick. • Take a look at the type definition for the object you’re passing. Make sure the 
modalities
property is defined in the type or interface you’re using. If it’s missing, you might need to extend the type or try a different configuration approach. • Also, it’s a good idea to review the LiveKit documentation or changelog for version 1.0.2 to see if 
modalities
is supported and if there are any special instructions for using it. If you’d like, I’m here to help you work through this further—just let me know!
b
which sdk version should i use can you send me npm link
tell
2 Views
Open in Slack
PreviousNext
Powered by