Allow 'conversational' summary_type with multichannel audio without requiring speaker_labels

[EnriqueRuvalcaba]

Allow 'conversational' summary_type with multichannel audio without requiring speaker_labels

Description

The 'conversational' summary type currently requires either speaker_labels or dual_channel to be enabled. However, multichannel is the recommended approach for handling multi-speaker audio, as dual_channel is deprecated.

Current Behavior

{
  multichannel: true,
  summary_type: 'conversational'
  // other options...
}

Throws error: "speaker_labels or dual_channel must be enabled for 'conversational' summaries"

Proposed Change

Update the validation logic to accept multichannel: true as a valid configuration for conversational summaries, since it provides the same speaker separation functionality as the deprecated dual_channel option.

[marcusolsson]

Thank you for letting us know! You're right that dual_channel is deprecated and should no longer be required.