Call logs say "User speech possibly starting" - but never really transcribes the audio recording. Call id: 8f4c8798-a610-4dd4-bee7-fbbd3061a468 https://cdn.discordapp.com/attachments/1351252649130987550/1351252649961328712/image.png?ex=67d9b398&is=67d86218&hm=5e9c2691f77b818a602fc93a56562f4d5deff22bc3557de412889e82154d328a&

Hey Bibek Shah, can you try another call with Deepgram transcriber settings: - Language: English - Pointing to 300ms - Model: Up to you (you can choose either Nova2 or Nova3) - Start speaking plan: Set smart endpointing to True - Start speaking plan:: Wait seconds to 0.6 These values will help the speech models to identify when the user has stopped speaking and it works most of the time and has been tested by most of the platform users. So give it a try and let me know how it goes for you.

Deepgram works yes, but for my use-case ElevenLabs transcriber seems to work properly..

Hey Bibek Shah from 11Labs Transcriber Issue. Can you share the call ID so that I can take a look? This will help me understand the problem you're facing.

d0da06a3-28b1-4acc-8c58-b1f2a64b9c86 09aba6e7-1a72-470f-8309-69a8b2524fee

Hello Bibek, it appears that 11labs failed during this call: 🔴 16:10:16:979 ElevenLabs Transcriber Error: pipeline-error-11labs-transcriber-failed - pipeline-error-11labs-transcriber-failed Could you please try again today and see if you're still running into this issue?