We're currently experiencing an issue where, despite the user clearly speaking (as visible in the orange waveform indicating user audio), the STT engine is not capturing or transcribing their speech. This has been happening consistently over the last hour and a half, regardless of the STT provider we use. We're seeing this even when the user's audio is properly recorded — it simply doesn't get transcribed, which is affecting critical parts of our flow, such as CPF validation via voice input.

Post @ #1211483291191083018

is it similar issue? https://discord.com/channels/1211482211119796234/1369082752493686794