Skip to main content

​
Timeline Entry

Each element under timeline contains multimodal signals for a single utterance.
​
utterance_id
string
Unique identifier for the utterance
​
speaker
string
Speaker identifier from diarization
​
t_start
number
Start time in seconds
​
t_end
number
End time in seconds
​
text
string
Transcribed text from speech recognition
​
asr_confidence
number
ASR confidence score (0-1)
​
prosody
object
Acoustic metrics including:
  • Pitch/energy statistics
  • Pauses
  • Spectral features
  • Commitment/passion markers
  • Lexical markers
  • Hesitation flags
​
prosody_desc
string
Natural-language descriptor (e.g., “COMMITTED, PASSIONATE, emphatic”)
​
emotion
object
Full emotion analysis including:
  • Emotion scores (joy, sadness, etc.)
  • PAD scores (valence, arousal, dominance)
  • Dominant emotion
  • Confidence
​
sentiment
object
Structured sentiment analysis
​
polarity
number
Sentiment polarity (-1 to 1)
​
subjectivity
number
Subjectivity score (0-1)
​
cognitive
object
Cognitive analysis including:
  • Engagement score
  • Cognitive load level
  • Cognitive signals
​
commitment_score
number
Derived commitment score from prosody
​
commitment_level
string
Commitment level classification
​
passion_score
number
Derived passion score from prosody
​
passion_level
string
Passion level classification
​
conviction_score
number
Conviction score derived from prosody and acoustic reasoning
​
lexical_markers
object
Lexical analysis including:
  • Hedges
  • Fillers
  • Questions
  • Disfluency rate
  • Self-repair flags
​
acoustic_reasoning
string
Narrative produced explaining how acoustic cues map to cognition/emotion
​
confidence
number
Average confidence across all agents (0-1)

​
ConversationSummary Schema (Reference)

Fields:
FieldTypeNotes
overviewstringExecutive-level synopsis.
key_insightsstring[]Bullet-ready highlights (timestamp references encouraged).
notable_patternsstring[]Trends across the conversation.
recommendationsstring[]Actionable next steps.
what_went_wellstring[] | nullPopulated when the user explicitly asks “what went well.”
what_went_wrongstring[] | nullPopulated on “what went wrong” requests.
speaker_dynamics[{speaker, description}] | nullCharacterization per speaker.
conversation_quality`“excellent""good""fair""poor”`Defaults to "good".
When contexts such as “call out what went well” are provided, those optional arrays will contain 3-5 evidence-backed bullets.

​
Insights

Aggregated insights from the fusion service.
​
overall_sentiment
object
Sentiment statistics:
  • mean: Average sentiment score
  • std: Standard deviation
  • trend: Sentiment trend (increasing/decreasing/stable)
​
dominant_emotions
array
Top three dominant emotions (strings)
​
engagement_pattern
string
Engagement pattern: increasing, decreasing, or stable
​
key_moments
array
Curated subset of detected moments (MomentEvent objects)
​
speaker_profiles
object
Profiles for each speaker
​
conversation_dynamics
object
Analysis of conversation dynamics
​
cognitive_profile
object
Overall cognitive profile

​
Transition Event

Affective state transition detection.
​
transition_id
string
Unique transition identifier
​
at_ms
integer
Timestamp in milliseconds
​
from_state
object
Previous affective state
​
to_state
object
New affective state
​
evidence_utterance_ids
array
Array of utterance IDs that contributed to this transition
​
drivers
object
Valence/arousal/engagement deltas that triggered the transition
​
confidence
number
Confidence score (0-1)
​
stability
string
Stability indicator

​
Moment Event

Key moment detection.
​
moment_id
string
Unique moment identifier
​
category
string
Moment category: objection, cta_offered, cta_accepted, cta_rejected, topic_shift
​
speaker_id
string
Speaker identifier
​
start_ms
integer
Start timestamp in milliseconds
​
end_ms
integer
End timestamp in milliseconds
​
utterance_ids
array
Array of associated utterance IDs
​
summary
string
Summary of the moment
​
labels
array
Additional labels
​
confidence
number
Confidence score (0-1)

​
Conversation Segmentation

Segmentation of conversation into phases.
​
phases
array
Array of ConversationPhase objects
​
total_phases
integer
Total number of phases
​
segmentation_rationale
string
Explanation of segmentation approach
​
narrative_arc
string
Narrative arc description
​
key_transitions
array
Array of key transitions (TransitionEvent objects)

​
Conversation Phase

Individual phase within a segmented conversation.
​
phase_id
string
Unique phase identifier
​
phase_name
string
Name of the phase
​
start_timestamp
number
Start timestamp in seconds
​
end_timestamp
number
End timestamp in seconds
​
start_utterance_id
string
First utterance ID in this phase
​
end_utterance_id
string
Last utterance ID in this phase
​
utterance_count
integer
Number of utterances in this phase
​
theme
string
Theme of the phase
​
summary
string
Summary of the phase
​
emotional_tone
string
Emotional tone of the phase
​
key_moments
array
Key moments within this phase (MomentEvent objects)
​
transition_signal
object
Transition signal that led to this phase (TransitionEvent)

​
Error Response

Standard error response format.
​
detail
string
Error message describing what went wrong
Example: 
{
  "detail": "Invalid or missing API key"
}

​
Live Session Response

Response from starting a live session.
​
session_id
string
Unique session identifier
​
expires_at
string
ISO 8601 timestamp when session expires
​
status
string
Current session status
​
next_chunk_seq
integer
Next expected chunk sequence number (zero-based)

​
Metrics Response

Observability metrics for dashboards.
​
events
object
Event counts by type:
  • status: Number of status events
  • final_transcript: Number of final transcript events
  • emotion: Number of emotion events
  • Additional event types as key-value pairs
​
agent_latency_ms
object
Average agent latency per utterance type (milliseconds)
Example: 
{
  "events": {
    "status": 45,
    "final_transcript": 12,
    "emotion": 12
  },
  "agent_latency_ms": {
    "utterance": 215.34
  }
}