This message was deleted.
# ask-ai
s
This message was deleted.
t
Thanks for reaching out, Dhruvil. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.
The issue you're encountering seems to be related to including timestamps like startTime and endTime for sentences in your session recording and transcripts. While the TranscriptionSegment interface in LiveKit does include properties for timestamps (such as startTime and endTime), the specific implementation details for including these timestamps in your transcripts are not explicitly documented. You may need to ensure that the transcription service or API you're using with LiveKit is configured to provide these timestamps. If you're using a custom transcription service, you might need to modify the service to include these details in the session history. If you can provide more details about your setup, such as the transcription service or API you're using, I can assist further or escalate this to a specialist for more in-depth support.
b
Can you provide me relevant documents link
t
Video or Audio Recording: Hey there! I totally get wanting to capture your session memories. LiveKit's Egress feature is like your personal recording buddy – it lets you easily record audio and video from your sessions. Imagine saving all those awesome moments by recording everything when participants are in the room. And the best part? You can store these recordings in cloud storage like Amazon S3, Google Cloud Storage, or Azure Blob Storage. If you want to see how it works, the documentation has some super helpful implementation details that'll walk you through it. • Text Transcripts with Timestamps: Here's something cool – LiveKit's transcription system is like having a super-smart assistant who not only captures what's said but also knows exactly when it was said! You'll get handy timestamps like
start_time_seconds
and
end_time_seconds
right in the API response. And even if timestamps aren't immediately obvious, the system is smart enough to estimate them based on audio buffer duration. Plus, there are built-in utilities to make sure those event timestamps are spot on. Want to dive deeper? The official LiveKit documentation is packed with great info, or feel free to ask me if you need me to point you to specific sections. I'm here to help! Sources: Session recording and transcripts | LiveKit Docs | GitHub::livekit::agents | livekit-plugins/livekit-plugins-sarvam/livekit/pl...