Class: Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent

Inherits:

Internal::Type::BaseModel

• Object
• Internal::Type::BaseModel
• Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent

Defined in:

lib/revox/models/call_create_response.rb,
sig/revox/models/call_create_response.rbs

Defined Under Namespace

Modules: Speaker

Instance Attribute Summary

• #id ⇒ String
• #is_real_turn ⇒ Boolean

  False when the segment was flagged as noise/blip rather than a meaningful turn.

• #llm_system_prompt ⇒ String^?

  Agent-only: despite the name, the whole prompt the LLM was given for this turn — system instructions plus conversation history — as JSON.

• #llm_ttft_ms ⇒ Float^?

  Agent-only: LLM time-to-first-token for the reply, in ms.

• #llm_winner_model ⇒ String^?

  Agent-only: the model that produced the reply — the winning leg when the LLM is a race, which is also the model llm_ttft_ms describes.

• #speaker ⇒ Symbol, Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent::Speaker

  Who was speaking during this segment.

• #started_speaking_at_ms ⇒ Float

  Milliseconds since session start when speech began.

• #stopped_speaking_at_ms ⇒ Float^?

  Milliseconds since session start when speech ended.

• #stt_transcription_delay_ms ⇒ Float^?

  User-only: time from end-of-speech to final transcript, in ms.

• #transcript ⇒ String^?

  Recognized text for this segment, when available.

• #tts_ttfb_ms ⇒ Float^?

  Agent-only: TTS time-to-first-audio-byte for the reply, in ms.

• #was_cut ⇒ Boolean^?

  Agent-only: true when the agent's audio was interrupted (real or auto-resumed).

Instance Method Summary

• #initialize(id:, is_real_turn:, llm_system_prompt:, llm_ttft_ms:, llm_winner_model:, speaker:, started_speaking_at_ms:, stopped_speaking_at_ms:, stt_transcription_delay_ms:, transcript:, tts_ttfb_ms:, was_cut:) ⇒ Object constructor

  Some parameter documentations has been truncated, see SpeechEvent for more details.

• #to_hash ⇒ {

Methods inherited from Internal::Type::BaseModel

==, #==, #[], coerce, #deconstruct_keys, #deep_to_h, dump, fields, hash, #hash, inherited, inspect, #inspect, known_fields, optional, recursively_to_h, required, #to_h, #to_json, #to_s, to_sorbet_type, #to_yaml

Methods included from Internal::Type::Converter

#coerce, coerce, #dump, dump, #inspect, inspect, meta_info, new_coerce_state, type_info

Methods included from Internal::Util::SorbetRuntimeSupport

#const_missing, #define_sorbet_constant!, #sorbet_constant_defined?, #to_sorbet_type, to_sorbet_type

Constructor Details

#initialize(id:, is_real_turn:, llm_system_prompt:, llm_ttft_ms:, llm_winner_model:, speaker:, started_speaking_at_ms:, stopped_speaking_at_ms:, stt_transcription_delay_ms:, transcript:, tts_ttfb_ms:, was_cut:) ⇒ Object

Some parameter documentations has been truncated, see Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent for more details.

Parameters:

• id (String)
• is_real_turn (Boolean) —

  False when the segment was flagged as noise/blip rather than a meaningful turn.

• llm_system_prompt (String, nil) —

  Agent-only: despite the name, the whole prompt the LLM was given for this turn —

• llm_ttft_ms (Float, nil) —

  Agent-only: LLM time-to-first-token for the reply, in ms.

• llm_winner_model (String, nil) —

  Agent-only: the model that produced the reply — the winning leg when the LLM is

• speaker (Symbol, Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent::Speaker) —

  Who was speaking during this segment.

• started_speaking_at_ms (Float) —

  Milliseconds since session start when speech began.

• stopped_speaking_at_ms (Float, nil) —

  Milliseconds since session start when speech ended. Null when the session ended

• stt_transcription_delay_ms (Float, nil) —

  User-only: time from end-of-speech to final transcript, in ms.

• transcript (String, nil) —

  Recognized text for this segment, when available.

• tts_ttfb_ms (Float, nil) —

  Agent-only: TTS time-to-first-audio-byte for the reply, in ms.

• was_cut (Boolean, nil) —

  Agent-only: true when the agent's audio was interrupted (real or auto-resumed).

# File 'lib/revox/models/call_create_response.rb', line 3390

Instance Attribute Details

#id ⇒ String

Parameters:

• value (String)

Returns:

• (String)

# File 'lib/revox/models/call_create_response.rb', line 3317

required :id, String

#is_real_turn ⇒ Boolean

False when the segment was flagged as noise/blip rather than a meaningful turn.

Parameters:

• value (Boolean)

Returns:

• (Boolean)

# File 'lib/revox/models/call_create_response.rb', line 3323

required :is_real_turn, Revox::Internal::Type::Boolean

#llm_system_prompt ⇒ String^?

Agent-only: despite the name, the whole prompt the LLM was given for this turn — system instructions plus conversation history — as JSON.

Parameters:

• value (String, nil)

Returns:

• (String, nil)

# File 'lib/revox/models/call_create_response.rb', line 3330

required :llm_system_prompt, String, nil?: true

#llm_ttft_ms ⇒ Float^?

Agent-only: LLM time-to-first-token for the reply, in ms.

Parameters:

• value (Float, nil)

Returns:

• (Float, nil)

# File 'lib/revox/models/call_create_response.rb', line 3336

required :llm_ttft_ms, Float, nil?: true

#llm_winner_model ⇒ String^?

Agent-only: the model that produced the reply — the winning leg when the LLM is a race, which is also the model llm_ttft_ms describes.

Parameters:

• value (String, nil)

Returns:

• (String, nil)

# File 'lib/revox/models/call_create_response.rb', line 3343

required :llm_winner_model, String, nil?: true

#speaker ⇒ Symbol, Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent::Speaker

Who was speaking during this segment.

Parameters:

• value (Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent::speaker)

Returns:

• (Symbol, Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent::Speaker)

# File 'lib/revox/models/call_create_response.rb', line 3349

required :speaker,
enum: -> { Revox::Models::CallCreateResponse::Call::LastCallAttempt::SpeechEvent::Speaker }

#started_speaking_at_ms ⇒ Float

Milliseconds since session start when speech began.

Parameters:

• value (Float)

Returns:

• (Float)

# File 'lib/revox/models/call_create_response.rb', line 3356

required :started_speaking_at_ms, Float

#stopped_speaking_at_ms ⇒ Float^?

Milliseconds since session start when speech ended. Null when the session ended before the turn closed.

Parameters:

• value (Float, nil)

Returns:

• (Float, nil)

# File 'lib/revox/models/call_create_response.rb', line 3363

required :stopped_speaking_at_ms, Float, nil?: true

#stt_transcription_delay_ms ⇒ Float^?

User-only: time from end-of-speech to final transcript, in ms.

Parameters:

• value (Float, nil)

Returns:

• (Float, nil)

# File 'lib/revox/models/call_create_response.rb', line 3369

required :stt_transcription_delay_ms, Float, nil?: true

#transcript ⇒ String^?

Recognized text for this segment, when available.

Parameters:

• value (String, nil)

Returns:

• (String, nil)

# File 'lib/revox/models/call_create_response.rb', line 3375

required :transcript, String, nil?: true

#tts_ttfb_ms ⇒ Float^?

Agent-only: TTS time-to-first-audio-byte for the reply, in ms.

Parameters:

• value (Float, nil)

Returns:

• (Float, nil)

# File 'lib/revox/models/call_create_response.rb', line 3381

required :tts_ttfb_ms, Float, nil?: true

#was_cut ⇒ Boolean^?

Agent-only: true when the agent's audio was interrupted (real or auto-resumed). Null on user segments.

Parameters:

• value (Boolean, nil)

Returns:

• (Boolean, nil)

# File 'lib/revox/models/call_create_response.rb', line 3388

required :was_cut, Revox::Internal::Type::Boolean, nil?: true

Instance Method Details

#to_hash ⇒ {

Returns:

• ({)

# File 'sig/revox/models/call_create_response.rbs', line 2737

def to_hash: -> {