On forward current version of the bot (in python) transcribes only first audio/voice message

[konard]

When multiple messages are forwarded, we need to support transcription of all them, not just a first one audio/voice message.