Skip to main content
GET
Get Call
Get detailed information about a specific call including transcript, turns, usage statistics, and cost breakdown.

Path Parameters

string
required
The unique identifier of the call (format: call_ prefix + alphanumeric).

Response

boolean
Indicates if the request was successful.
object
Complete call data object.

Call Data Fields

string
required
Unique call identifier.
string
required
Phone number that was called (E.164 format).
string
Caller phone number.
string
required
ID of the agent that handled the call.
string
Name of the agent.
object
Agent configuration used during the call (LLM model, TTS provider, temperature, etc.).
string
Type of call: inbound, outbound, or preview.
string
Current call status.
string
Transport method: twilio, sip, or webrtc.
boolean
Whether recording was enabled.
string
S3 URL to the processed recording file.
number
Call duration in seconds.
number
Number of conversation turns.
number
Total cost in credits.
object
Detailed cost breakdown by component:
  • llm_input: Input token costs
  • llm_output: Output token costs
  • tts: Text-to-speech costs
  • stt: Speech-to-text costs
  • telephony: Telephony costs
object
Usage statistics:
  • llm: Token usage (input_tokens, output_tokens, total_tokens)
  • tts: Character count
  • stt: Audio duration in seconds
object
Call summary metrics:
  • end_reason: Why the call ended
  • interruptions: Number of interruptions
  • avg_turn_duration: Average turn duration in seconds
  • total_tokens: Total LLM tokens used
  • total_characters: Total TTS characters
object
Error information if any errors occurred:
  • error_count: Number of errors
  • has_fatal_errors: Whether any errors were fatal
  • errors: Array of error details
array
Array of conversation turns with detailed events.
array
Array of transcript entries with speaker, text, and timestamps.
string
ISO8601 timestamp when the call was created.
string
ISO8601 timestamp when the call started.
string
ISO8601 timestamp when the call ended.
The turns array contains detailed conversation events including tool calls, LLM metadata, TTS metadata, and STT metadata. The transcript array provides a simplified view with speaker and text for each turn.