openai 0.22.1 → 0.23.0

data/lib/openai/models/realtime/transcription_session_created.rb

@@ -11,10 +11,14 @@ module OpenAI
         required :event_id, String
 
         # @!attribute session
-        #   A Realtime transcription session configuration object.
+        #   A new Realtime transcription session configuration.
         #
-        #   @return [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session]
-        required :session, -> { OpenAI::Realtime::TranscriptionSessionCreated::Session }
+        #   When a session is created on the server via REST API, the session object also
+        #   contains an ephemeral key. Default TTL for keys is 10 minutes. This property is
+        #   not present when a session is updated via the WebSocket API.
+        #
+        #   @return [OpenAI::Models::Realtime::RealtimeTranscriptionSessionCreateResponse]
+        required :session, -> { OpenAI::Realtime::RealtimeTranscriptionSessionCreateResponse }
 
         # @!attribute type
         #   The event type, must be `transcription_session.created`.
@@ -30,248 +34,9 @@ module OpenAI
         #
         #   @param event_id [String] The unique ID of the server event.
         #
-        #   @param session [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session] A Realtime transcription session configuration object.
+        #   @param session [OpenAI::Models::Realtime::RealtimeTranscriptionSessionCreateResponse] A new Realtime transcription session configuration.
         #
         #   @param type [Symbol, :"transcription_session.created"] The event type, must be `transcription_session.created`.
-
-        # @see OpenAI::Models::Realtime::TranscriptionSessionCreated#session
-        class Session < OpenAI::Internal::Type::BaseModel
-          # @!attribute id
-          #   Unique identifier for the session that looks like `sess_1234567890abcdef`.
-          #
-          #   @return [String, nil]
-          optional :id, String
-
-          # @!attribute audio
-          #   Configuration for input audio for the session.
-          #
-          #   @return [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio, nil]
-          optional :audio, -> { OpenAI::Realtime::TranscriptionSessionCreated::Session::Audio }
-
-          # @!attribute expires_at
-          #   Expiration timestamp for the session, in seconds since epoch.
-          #
-          #   @return [Integer, nil]
-          optional :expires_at, Integer
-
-          # @!attribute include
-          #   Additional fields to include in server outputs.
-          #
-          #   - `item.input_audio_transcription.logprobs`: Include logprobs for input audio
-          #     transcription.
-          #
-          #   @return [Array<Symbol, OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Include>, nil]
-          optional :include,
-                   -> { OpenAI::Internal::Type::ArrayOf[enum: OpenAI::Realtime::TranscriptionSessionCreated::Session::Include] }
-
-          # @!attribute object
-          #   The object type. Always `realtime.transcription_session`.
-          #
-          #   @return [String, nil]
-          optional :object, String
-
-          # @!method initialize(id: nil, audio: nil, expires_at: nil, include: nil, object: nil)
-          #   Some parameter documentations has been truncated, see
-          #   {OpenAI::Models::Realtime::TranscriptionSessionCreated::Session} for more
-          #   details.
-          #
-          #   A Realtime transcription session configuration object.
-          #
-          #   @param id [String] Unique identifier for the session that looks like `sess_1234567890abcdef`.
-          #
-          #   @param audio [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio] Configuration for input audio for the session.
-          #
-          #   @param expires_at [Integer] Expiration timestamp for the session, in seconds since epoch.
-          #
-          #   @param include [Array<Symbol, OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Include>] Additional fields to include in server outputs.
-          #
-          #   @param object [String] The object type. Always `realtime.transcription_session`.
-
-          # @see OpenAI::Models::Realtime::TranscriptionSessionCreated::Session#audio
-          class Audio < OpenAI::Internal::Type::BaseModel
-            # @!attribute input
-            #
-            #   @return [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input, nil]
-            optional :input, -> { OpenAI::Realtime::TranscriptionSessionCreated::Session::Audio::Input }
-
-            # @!method initialize(input: nil)
-            #   Configuration for input audio for the session.
-            #
-            #   @param input [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input]
-
-            # @see OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio#input
-            class Input < OpenAI::Internal::Type::BaseModel
-              # @!attribute format_
-              #   The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
-              #
-              #   @return [String, nil]
-              optional :format_, String, api_name: :format
-
-              # @!attribute noise_reduction
-              #   Configuration for input audio noise reduction.
-              #
-              #   @return [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::NoiseReduction, nil]
-              optional :noise_reduction,
-                       -> { OpenAI::Realtime::TranscriptionSessionCreated::Session::Audio::Input::NoiseReduction }
-
-              # @!attribute transcription
-              #   Configuration of the transcription model.
-              #
-              #   @return [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::Transcription, nil]
-              optional :transcription,
-                       -> { OpenAI::Realtime::TranscriptionSessionCreated::Session::Audio::Input::Transcription }
-
-              # @!attribute turn_detection
-              #   Configuration for turn detection.
-              #
-              #   @return [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::TurnDetection, nil]
-              optional :turn_detection,
-                       -> { OpenAI::Realtime::TranscriptionSessionCreated::Session::Audio::Input::TurnDetection }
-
-              # @!method initialize(format_: nil, noise_reduction: nil, transcription: nil, turn_detection: nil)
-              #   Some parameter documentations has been truncated, see
-              #   {OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input}
-              #   for more details.
-              #
-              #   @param format_ [String] The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
-              #
-              #   @param noise_reduction [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::NoiseReduction] Configuration for input audio noise reduction.
-              #
-              #   @param transcription [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::Transcription] Configuration of the transcription model.
-              #
-              #   @param turn_detection [OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::TurnDetection] Configuration for turn detection.
-
-              # @see OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input#noise_reduction
-              class NoiseReduction < OpenAI::Internal::Type::BaseModel
-                # @!attribute type
-                #
-                #   @return [Symbol, OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::NoiseReduction::Type, nil]
-                optional :type,
-                         enum: -> { OpenAI::Realtime::TranscriptionSessionCreated::Session::Audio::Input::NoiseReduction::Type }
-
-                # @!method initialize(type: nil)
-                #   Configuration for input audio noise reduction.
-                #
-                #   @param type [Symbol, OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::NoiseReduction::Type]
-
-                # @see OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::NoiseReduction#type
-                module Type
-                  extend OpenAI::Internal::Type::Enum
-
-                  NEAR_FIELD = :near_field
-                  FAR_FIELD = :far_field
-
-                  # @!method self.values
-                  #   @return [Array<Symbol>]
-                end
-              end
-
-              # @see OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input#transcription
-              class Transcription < OpenAI::Internal::Type::BaseModel
-                # @!attribute language
-                #   The language of the input audio. Supplying the input language in
-                #   [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`)
-                #   format will improve accuracy and latency.
-                #
-                #   @return [String, nil]
-                optional :language, String
-
-                # @!attribute model
-                #   The model to use for transcription. Can be `gpt-4o-transcribe`,
-                #   `gpt-4o-mini-transcribe`, or `whisper-1`.
-                #
-                #   @return [Symbol, OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::Transcription::Model, nil]
-                optional :model,
-                         enum: -> { OpenAI::Realtime::TranscriptionSessionCreated::Session::Audio::Input::Transcription::Model }
-
-                # @!attribute prompt
-                #   An optional text to guide the model's style or continue a previous audio
-                #   segment. The
-                #   [prompt](https://platform.openai.com/docs/guides/speech-to-text#prompting)
-                #   should match the audio language.
-                #
-                #   @return [String, nil]
-                optional :prompt, String
-
-                # @!method initialize(language: nil, model: nil, prompt: nil)
-                #   Some parameter documentations has been truncated, see
-                #   {OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::Transcription}
-                #   for more details.
-                #
-                #   Configuration of the transcription model.
-                #
-                #   @param language [String] The language of the input audio. Supplying the input language in
-                #
-                #   @param model [Symbol, OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::Transcription::Model] The model to use for transcription. Can be `gpt-4o-transcribe`, `gpt-4o-mini-tra
-                #
-                #   @param prompt [String] An optional text to guide the model's style or continue a previous audio segment
-
-                # The model to use for transcription. Can be `gpt-4o-transcribe`,
-                # `gpt-4o-mini-transcribe`, or `whisper-1`.
-                #
-                # @see OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::Transcription#model
-                module Model
-                  extend OpenAI::Internal::Type::Enum
-
-                  GPT_4O_TRANSCRIBE = :"gpt-4o-transcribe"
-                  GPT_4O_MINI_TRANSCRIBE = :"gpt-4o-mini-transcribe"
-                  WHISPER_1 = :"whisper-1"
-
-                  # @!method self.values
-                  #   @return [Array<Symbol>]
-                end
-              end
-
-              # @see OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input#turn_detection
-              class TurnDetection < OpenAI::Internal::Type::BaseModel
-                # @!attribute prefix_padding_ms
-                #
-                #   @return [Integer, nil]
-                optional :prefix_padding_ms, Integer
-
-                # @!attribute silence_duration_ms
-                #
-                #   @return [Integer, nil]
-                optional :silence_duration_ms, Integer
-
-                # @!attribute threshold
-                #
-                #   @return [Float, nil]
-                optional :threshold, Float
-
-                # @!attribute type
-                #   Type of turn detection, only `server_vad` is currently supported.
-                #
-                #   @return [String, nil]
-                optional :type, String
-
-                # @!method initialize(prefix_padding_ms: nil, silence_duration_ms: nil, threshold: nil, type: nil)
-                #   Some parameter documentations has been truncated, see
-                #   {OpenAI::Models::Realtime::TranscriptionSessionCreated::Session::Audio::Input::TurnDetection}
-                #   for more details.
-                #
-                #   Configuration for turn detection.
-                #
-                #   @param prefix_padding_ms [Integer]
-                #
-                #   @param silence_duration_ms [Integer]
-                #
-                #   @param threshold [Float]
-                #
-                #   @param type [String] Type of turn detection, only `server_vad` is currently supported.
-              end
-            end
-          end
-
-          module Include
-            extend OpenAI::Internal::Type::Enum
-
-            ITEM_INPUT_AUDIO_TRANSCRIPTION_LOGPROBS = :"item.input_audio_transcription.logprobs"
-
-            # @!method self.values
-            #   @return [Array<Symbol>]
-          end
-        end
       end
     end
   end

data/lib/openai/models/realtime/transcription_session_update.rb

@@ -7,8 +7,8 @@ module OpenAI
         # @!attribute session
         #   Realtime transcription session object configuration.
         #
-        #   @return [OpenAI::Models::Realtime::RealtimeTranscriptionSessionCreateRequest]
-        required :session, -> { OpenAI::Realtime::RealtimeTranscriptionSessionCreateRequest }
+        #   @return [OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session]
+        required :session, -> { OpenAI::Realtime::TranscriptionSessionUpdate::Session }
 
         # @!attribute type
         #   The event type, must be `transcription_session.update`.
@@ -25,11 +25,187 @@ module OpenAI
         # @!method initialize(session:, event_id: nil, type: :"transcription_session.update")
         #   Send this event to update a transcription session.
         #
-        #   @param session [OpenAI::Models::Realtime::RealtimeTranscriptionSessionCreateRequest] Realtime transcription session object configuration.
+        #   @param session [OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session] Realtime transcription session object configuration.
         #
         #   @param event_id [String] Optional client-generated ID used to identify this event.
         #
         #   @param type [Symbol, :"transcription_session.update"] The event type, must be `transcription_session.update`.
+
+        # @see OpenAI::Models::Realtime::TranscriptionSessionUpdate#session
+        class Session < OpenAI::Internal::Type::BaseModel
+          # @!attribute include
+          #   The set of items to include in the transcription. Current available items are:
+          #   `item.input_audio_transcription.logprobs`
+          #
+          #   @return [Array<Symbol, OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::Include>, nil]
+          optional :include,
+                   -> { OpenAI::Internal::Type::ArrayOf[enum: OpenAI::Realtime::TranscriptionSessionUpdate::Session::Include] }
+
+          # @!attribute input_audio_format
+          #   The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. For
+          #   `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate, single channel
+          #   (mono), and little-endian byte order.
+          #
+          #   @return [Symbol, OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::InputAudioFormat, nil]
+          optional :input_audio_format,
+                   enum: -> { OpenAI::Realtime::TranscriptionSessionUpdate::Session::InputAudioFormat }
+
+          # @!attribute input_audio_noise_reduction
+          #   Configuration for input audio noise reduction. This can be set to `null` to turn
+          #   off. Noise reduction filters audio added to the input audio buffer before it is
+          #   sent to VAD and the model. Filtering the audio can improve VAD and turn
+          #   detection accuracy (reducing false positives) and model performance by improving
+          #   perception of the input audio.
+          #
+          #   @return [OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::InputAudioNoiseReduction, nil]
+          optional :input_audio_noise_reduction,
+                   -> { OpenAI::Realtime::TranscriptionSessionUpdate::Session::InputAudioNoiseReduction }
+
+          # @!attribute input_audio_transcription
+          #   Configuration for input audio transcription. The client can optionally set the
+          #   language and prompt for transcription, these offer additional guidance to the
+          #   transcription service.
+          #
+          #   @return [OpenAI::Models::Realtime::AudioTranscription, nil]
+          optional :input_audio_transcription, -> { OpenAI::Realtime::AudioTranscription }
+
+          # @!attribute turn_detection
+          #   Configuration for turn detection. Can be set to `null` to turn off. Server VAD
+          #   means that the model will detect the start and end of speech based on audio
+          #   volume and respond at the end of user speech.
+          #
+          #   @return [OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::TurnDetection, nil]
+          optional :turn_detection, -> { OpenAI::Realtime::TranscriptionSessionUpdate::Session::TurnDetection }
+
+          # @!method initialize(include: nil, input_audio_format: nil, input_audio_noise_reduction: nil, input_audio_transcription: nil, turn_detection: nil)
+          #   Some parameter documentations has been truncated, see
+          #   {OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session} for more
+          #   details.
+          #
+          #   Realtime transcription session object configuration.
+          #
+          #   @param include [Array<Symbol, OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::Include>] The set of items to include in the transcription. Current available items are:
+          #
+          #   @param input_audio_format [Symbol, OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::InputAudioFormat] The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
+          #
+          #   @param input_audio_noise_reduction [OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::InputAudioNoiseReduction] Configuration for input audio noise reduction. This can be set to `null` to turn
+          #
+          #   @param input_audio_transcription [OpenAI::Models::Realtime::AudioTranscription] Configuration for input audio transcription. The client can optionally set the l
+          #
+          #   @param turn_detection [OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::TurnDetection] Configuration for turn detection. Can be set to `null` to turn off. Server VAD m
+
+          module Include
+            extend OpenAI::Internal::Type::Enum
+
+            ITEM_INPUT_AUDIO_TRANSCRIPTION_LOGPROBS = :"item.input_audio_transcription.logprobs"
+
+            # @!method self.values
+            #   @return [Array<Symbol>]
+          end
+
+          # The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. For
+          # `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate, single channel
+          # (mono), and little-endian byte order.
+          #
+          # @see OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session#input_audio_format
+          module InputAudioFormat
+            extend OpenAI::Internal::Type::Enum
+
+            PCM16 = :pcm16
+            G711_ULAW = :g711_ulaw
+            G711_ALAW = :g711_alaw
+
+            # @!method self.values
+            #   @return [Array<Symbol>]
+          end
+
+          # @see OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session#input_audio_noise_reduction
+          class InputAudioNoiseReduction < OpenAI::Internal::Type::BaseModel
+            # @!attribute type
+            #   Type of noise reduction. `near_field` is for close-talking microphones such as
+            #   headphones, `far_field` is for far-field microphones such as laptop or
+            #   conference room microphones.
+            #
+            #   @return [Symbol, OpenAI::Models::Realtime::NoiseReductionType, nil]
+            optional :type, enum: -> { OpenAI::Realtime::NoiseReductionType }
+
+            # @!method initialize(type: nil)
+            #   Some parameter documentations has been truncated, see
+            #   {OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::InputAudioNoiseReduction}
+            #   for more details.
+            #
+            #   Configuration for input audio noise reduction. This can be set to `null` to turn
+            #   off. Noise reduction filters audio added to the input audio buffer before it is
+            #   sent to VAD and the model. Filtering the audio can improve VAD and turn
+            #   detection accuracy (reducing false positives) and model performance by improving
+            #   perception of the input audio.
+            #
+            #   @param type [Symbol, OpenAI::Models::Realtime::NoiseReductionType] Type of noise reduction. `near_field` is for close-talking microphones such as h
+          end
+
+          # @see OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session#turn_detection
+          class TurnDetection < OpenAI::Internal::Type::BaseModel
+            # @!attribute prefix_padding_ms
+            #   Amount of audio to include before the VAD detected speech (in milliseconds).
+            #   Defaults to 300ms.
+            #
+            #   @return [Integer, nil]
+            optional :prefix_padding_ms, Integer
+
+            # @!attribute silence_duration_ms
+            #   Duration of silence to detect speech stop (in milliseconds). Defaults to 500ms.
+            #   With shorter values the model will respond more quickly, but may jump in on
+            #   short pauses from the user.
+            #
+            #   @return [Integer, nil]
+            optional :silence_duration_ms, Integer
+
+            # @!attribute threshold
+            #   Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A higher
+            #   threshold will require louder audio to activate the model, and thus might
+            #   perform better in noisy environments.
+            #
+            #   @return [Float, nil]
+            optional :threshold, Float
+
+            # @!attribute type
+            #   Type of turn detection. Only `server_vad` is currently supported for
+            #   transcription sessions.
+            #
+            #   @return [Symbol, OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::TurnDetection::Type, nil]
+            optional :type, enum: -> { OpenAI::Realtime::TranscriptionSessionUpdate::Session::TurnDetection::Type }
+
+            # @!method initialize(prefix_padding_ms: nil, silence_duration_ms: nil, threshold: nil, type: nil)
+            #   Some parameter documentations has been truncated, see
+            #   {OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::TurnDetection}
+            #   for more details.
+            #
+            #   Configuration for turn detection. Can be set to `null` to turn off. Server VAD
+            #   means that the model will detect the start and end of speech based on audio
+            #   volume and respond at the end of user speech.
+            #
+            #   @param prefix_padding_ms [Integer] Amount of audio to include before the VAD detected speech (in
+            #
+            #   @param silence_duration_ms [Integer] Duration of silence to detect speech stop (in milliseconds). Defaults
+            #
+            #   @param threshold [Float] Activation threshold for VAD (0.0 to 1.0), this defaults to 0.5. A
+            #
+            #   @param type [Symbol, OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::TurnDetection::Type] Type of turn detection. Only `server_vad` is currently supported for transcripti
+
+            # Type of turn detection. Only `server_vad` is currently supported for
+            # transcription sessions.
+            #
+            # @see OpenAI::Models::Realtime::TranscriptionSessionUpdate::Session::TurnDetection#type
+            module Type
+              extend OpenAI::Internal::Type::Enum
+
+              SERVER_VAD = :server_vad
+
+              # @!method self.values
+              #   @return [Array<Symbol>]
+            end
+          end
+        end
       end
     end
   end