The performed cues in a call transcript line, in order, without brackets
(0.70.0): '[laughs] 好呀 [whispers] 秘密' → ['laughs', 'whispers']. The
same cues stripAudioTags removes, so the two always agree. connectCall
reports them on each companion line of a call where audioTags was
granted (onTranscript's audioTags); exported for self-plumbed
ElevenLabs sessions.
The performed cues in a call transcript line, in order, without brackets (0.70.0):
'[laughs] 好呀 [whispers] 秘密'→['laughs', 'whispers']. The same cuesstripAudioTagsremoves, so the two always agree.connectCallreports them on each companion line of a call whereaudioTagswas granted (onTranscript'saudioTags); exported for self-plumbed ElevenLabs sessions.