openai 0.31.0 → 0.32.0

data/rbi/openai/models/audio/transcription_diarized_segment.rbi

@@ -0,0 +1,87 @@
+# typed: strong
+
+module OpenAI
+  module Models
+    module Audio
+      class TranscriptionDiarizedSegment < OpenAI::Internal::Type::BaseModel
+        OrHash =
+          T.type_alias do
+            T.any(
+              OpenAI::Audio::TranscriptionDiarizedSegment,
+              OpenAI::Internal::AnyHash
+            )
+          end
+
+        # Unique identifier for the segment.
+        sig { returns(String) }
+        attr_accessor :id
+
+        # End timestamp of the segment in seconds.
+        sig { returns(Float) }
+        attr_accessor :end_
+
+        # Speaker label for this segment. When known speakers are provided, the label
+        # matches `known_speaker_names[]`. Otherwise speakers are labeled sequentially
+        # using capital letters (`A`, `B`, ...).
+        sig { returns(String) }
+        attr_accessor :speaker
+
+        # Start timestamp of the segment in seconds.
+        sig { returns(Float) }
+        attr_accessor :start
+
+        # Transcript text for this segment.
+        sig { returns(String) }
+        attr_accessor :text
+
+        # The type of the segment. Always `transcript.text.segment`.
+        sig { returns(Symbol) }
+        attr_accessor :type
+
+        # A segment of diarized transcript text with speaker metadata.
+        sig do
+          params(
+            id: String,
+            end_: Float,
+            speaker: String,
+            start: Float,
+            text: String,
+            type: Symbol
+          ).returns(T.attached_class)
+        end
+        def self.new(
+          # Unique identifier for the segment.
+          id:,
+          # End timestamp of the segment in seconds.
+          end_:,
+          # Speaker label for this segment. When known speakers are provided, the label
+          # matches `known_speaker_names[]`. Otherwise speakers are labeled sequentially
+          # using capital letters (`A`, `B`, ...).
+          speaker:,
+          # Start timestamp of the segment in seconds.
+          start:,
+          # Transcript text for this segment.
+          text:,
+          # The type of the segment. Always `transcript.text.segment`.
+          type: :"transcript.text.segment"
+        )
+        end
+
+        sig do
+          override.returns(
+            {
+              id: String,
+              end_: Float,
+              speaker: String,
+              start: Float,
+              text: String,
+              type: Symbol
+            }
+          )
+        end
+        def to_hash
+        end
+      end
+    end
+  end
+end

data/rbi/openai/models/audio/transcription_stream_event.rbi

@@ -3,16 +3,17 @@
 module OpenAI
   module Models
     module Audio
-      # Emitted when there is an additional text delta. This is also the first event
-      # emitted when the transcription starts. Only emitted when you
+      # Emitted when a diarized transcription returns a completed segment with speaker
+      # information. Only emitted when you
       # [create a transcription](https://platform.openai.com/docs/api-reference/audio/create-transcription)
-      # with the `Stream` parameter set to `true`.
+      # with `stream` set to `true` and `response_format` set to `diarized_json`.
       module TranscriptionStreamEvent
         extend OpenAI::Internal::Type::Union
 
         Variants =
           T.type_alias do
             T.any(
+              OpenAI::Audio::TranscriptionTextSegmentEvent,
               OpenAI::Audio::TranscriptionTextDeltaEvent,
               OpenAI::Audio::TranscriptionTextDoneEvent
             )

data/rbi/openai/models/audio/transcription_text_delta_event.rbi

@@ -42,6 +42,14 @@ module OpenAI
         end
         attr_writer :logprobs
 
+        # Identifier of the diarized segment that this delta belongs to. Only present when
+        # using `gpt-4o-transcribe-diarize`.
+        sig { returns(T.nilable(String)) }
+        attr_reader :segment_id
+
+        sig { params(segment_id: String).void }
+        attr_writer :segment_id
+
         # Emitted when there is an additional text delta. This is also the first event
         # emitted when the transcription starts. Only emitted when you
         # [create a transcription](https://platform.openai.com/docs/api-reference/audio/create-transcription)
@@ -53,6 +61,7 @@ module OpenAI
               T::Array[
                 OpenAI::Audio::TranscriptionTextDeltaEvent::Logprob::OrHash
               ],
+            segment_id: String,
             type: Symbol
           ).returns(T.attached_class)
         end
@@ -63,6 +72,9 @@ module OpenAI
           # [create a transcription](https://platform.openai.com/docs/api-reference/audio/create-transcription)
           # with the `include[]` parameter set to `logprobs`.
           logprobs: nil,
+          # Identifier of the diarized segment that this delta belongs to. Only present when
+          # using `gpt-4o-transcribe-diarize`.
+          segment_id: nil,
           # The type of the event. Always `transcript.text.delta`.
           type: :"transcript.text.delta"
         )
@@ -74,7 +86,8 @@ module OpenAI
               delta: String,
               type: Symbol,
               logprobs:
-                T::Array[OpenAI::Audio::TranscriptionTextDeltaEvent::Logprob]
+                T::Array[OpenAI::Audio::TranscriptionTextDeltaEvent::Logprob],
+              segment_id: String
             }
           )
         end

data/rbi/openai/models/audio/transcription_text_segment_event.rbi

@@ -0,0 +1,86 @@
+# typed: strong
+
+module OpenAI
+  module Models
+    module Audio
+      class TranscriptionTextSegmentEvent < OpenAI::Internal::Type::BaseModel
+        OrHash =
+          T.type_alias do
+            T.any(
+              OpenAI::Audio::TranscriptionTextSegmentEvent,
+              OpenAI::Internal::AnyHash
+            )
+          end
+
+        # Unique identifier for the segment.
+        sig { returns(String) }
+        attr_accessor :id
+
+        # End timestamp of the segment in seconds.
+        sig { returns(Float) }
+        attr_accessor :end_
+
+        # Speaker label for this segment.
+        sig { returns(String) }
+        attr_accessor :speaker
+
+        # Start timestamp of the segment in seconds.
+        sig { returns(Float) }
+        attr_accessor :start
+
+        # Transcript text for this segment.
+        sig { returns(String) }
+        attr_accessor :text
+
+        # The type of the event. Always `transcript.text.segment`.
+        sig { returns(Symbol) }
+        attr_accessor :type
+
+        # Emitted when a diarized transcription returns a completed segment with speaker
+        # information. Only emitted when you
+        # [create a transcription](https://platform.openai.com/docs/api-reference/audio/create-transcription)
+        # with `stream` set to `true` and `response_format` set to `diarized_json`.
+        sig do
+          params(
+            id: String,
+            end_: Float,
+            speaker: String,
+            start: Float,
+            text: String,
+            type: Symbol
+          ).returns(T.attached_class)
+        end
+        def self.new(
+          # Unique identifier for the segment.
+          id:,
+          # End timestamp of the segment in seconds.
+          end_:,
+          # Speaker label for this segment.
+          speaker:,
+          # Start timestamp of the segment in seconds.
+          start:,
+          # Transcript text for this segment.
+          text:,
+          # The type of the event. Always `transcript.text.segment`.
+          type: :"transcript.text.segment"
+        )
+        end
+
+        sig do
+          override.returns(
+            {
+              id: String,
+              end_: Float,
+              speaker: String,
+              start: Float,
+              text: String,
+              type: Symbol
+            }
+          )
+        end
+        def to_hash
+        end
+      end
+    end
+  end
+end

data/rbi/openai/models/audio_model.rbi

@@ -13,6 +13,8 @@ module OpenAI
         T.let(:"gpt-4o-transcribe", OpenAI::AudioModel::TaggedSymbol)
       GPT_4O_MINI_TRANSCRIBE =
         T.let(:"gpt-4o-mini-transcribe", OpenAI::AudioModel::TaggedSymbol)
+      GPT_4O_TRANSCRIBE_DIARIZE =
+        T.let(:"gpt-4o-transcribe-diarize", OpenAI::AudioModel::TaggedSymbol)
 
       sig { override.returns(T::Array[OpenAI::AudioModel::TaggedSymbol]) }
       def self.values

data/rbi/openai/models/audio_response_format.rbi

@@ -3,8 +3,10 @@
 module OpenAI
   module Models
     # The format of the output, in one of these options: `json`, `text`, `srt`,
-    # `verbose_json`, or `vtt`. For `gpt-4o-transcribe` and `gpt-4o-mini-transcribe`,
-    # the only supported format is `json`.
+    # `verbose_json`, `vtt`, or `diarized_json`. For `gpt-4o-transcribe` and
+    # `gpt-4o-mini-transcribe`, the only supported format is `json`. For
+    # `gpt-4o-transcribe-diarize`, the supported formats are `json`, `text`, and
+    # `diarized_json`, with `diarized_json` required to receive speaker annotations.
     module AudioResponseFormat
       extend OpenAI::Internal::Type::Enum
 
@@ -17,6 +19,8 @@ module OpenAI
       VERBOSE_JSON =
         T.let(:verbose_json, OpenAI::AudioResponseFormat::TaggedSymbol)
       VTT = T.let(:vtt, OpenAI::AudioResponseFormat::TaggedSymbol)
+      DIARIZED_JSON =
+        T.let(:diarized_json, OpenAI::AudioResponseFormat::TaggedSymbol)
 
       sig do
         override.returns(T::Array[OpenAI::AudioResponseFormat::TaggedSymbol])

data/rbi/openai/models/realtime/audio_transcription.rbi

@@ -22,7 +22,8 @@ module OpenAI
         attr_writer :language
 
         # The model to use for transcription. Current options are `whisper-1`,
-        # `gpt-4o-transcribe-latest`, `gpt-4o-mini-transcribe`, and `gpt-4o-transcribe`.
+        # `gpt-4o-mini-transcribe`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`.
+        # Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels.
         sig do
           returns(
             T.nilable(OpenAI::Realtime::AudioTranscription::Model::OrSymbol)
@@ -40,8 +41,8 @@ module OpenAI
         # An optional text to guide the model's style or continue a previous audio
         # segment. For `whisper-1`, the
         # [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting).
-        # For `gpt-4o-transcribe` models, the prompt is a free text string, for example
-        # "expect words related to technology".
+        # For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the
+        # prompt is a free text string, for example "expect words related to technology".
         sig { returns(T.nilable(String)) }
         attr_reader :prompt
 
@@ -61,13 +62,14 @@ module OpenAI
           # format will improve accuracy and latency.
           language: nil,
           # The model to use for transcription. Current options are `whisper-1`,
-          # `gpt-4o-transcribe-latest`, `gpt-4o-mini-transcribe`, and `gpt-4o-transcribe`.
+          # `gpt-4o-mini-transcribe`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`.
+          # Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels.
           model: nil,
           # An optional text to guide the model's style or continue a previous audio
           # segment. For `whisper-1`, the
           # [prompt is a list of keywords](https://platform.openai.com/docs/guides/speech-to-text#prompting).
-          # For `gpt-4o-transcribe` models, the prompt is a free text string, for example
-          # "expect words related to technology".
+          # For `gpt-4o-transcribe` models (excluding `gpt-4o-transcribe-diarize`), the
+          # prompt is a free text string, for example "expect words related to technology".
           prompt: nil
         )
         end
@@ -85,7 +87,8 @@ module OpenAI
         end
 
         # The model to use for transcription. Current options are `whisper-1`,
-        # `gpt-4o-transcribe-latest`, `gpt-4o-mini-transcribe`, and `gpt-4o-transcribe`.
+        # `gpt-4o-mini-transcribe`, `gpt-4o-transcribe`, and `gpt-4o-transcribe-diarize`.
+        # Use `gpt-4o-transcribe-diarize` when you need diarization with speaker labels.
         module Model
           extend OpenAI::Internal::Type::Enum
 
@@ -100,11 +103,6 @@ module OpenAI
               :"whisper-1",
               OpenAI::Realtime::AudioTranscription::Model::TaggedSymbol
             )
-          GPT_4O_TRANSCRIBE_LATEST =
-            T.let(
-              :"gpt-4o-transcribe-latest",
-              OpenAI::Realtime::AudioTranscription::Model::TaggedSymbol
-            )
           GPT_4O_MINI_TRANSCRIBE =
             T.let(
               :"gpt-4o-mini-transcribe",
@@ -115,6 +113,11 @@ module OpenAI
               :"gpt-4o-transcribe",
               OpenAI::Realtime::AudioTranscription::Model::TaggedSymbol
             )
+          GPT_4O_TRANSCRIBE_DIARIZE =
+            T.let(
+              :"gpt-4o-transcribe-diarize",
+              OpenAI::Realtime::AudioTranscription::Model::TaggedSymbol
+            )
 
           sig do
             override.returns(

data/rbi/openai/models/vector_store_create_params.rbi

@@ -36,6 +36,14 @@ module OpenAI
       end
       attr_writer :chunking_strategy
 
+      # A description for the vector store. Can be used to describe the vector store's
+      # purpose.
+      sig { returns(T.nilable(String)) }
+      attr_reader :description
+
+      sig { params(description: String).void }
+      attr_writer :description
+
       # The expiration policy for a vector store.
       sig { returns(T.nilable(OpenAI::VectorStoreCreateParams::ExpiresAfter)) }
       attr_reader :expires_after
@@ -79,6 +87,7 @@ module OpenAI
               OpenAI::AutoFileChunkingStrategyParam::OrHash,
               OpenAI::StaticFileChunkingStrategyObjectParam::OrHash
             ),
+          description: String,
           expires_after: OpenAI::VectorStoreCreateParams::ExpiresAfter::OrHash,
           file_ids: T::Array[String],
           metadata: T.nilable(T::Hash[Symbol, String]),
@@ -90,6 +99,9 @@ module OpenAI
         # The chunking strategy used to chunk the file(s). If not set, will use the `auto`
         # strategy. Only applicable if `file_ids` is non-empty.
         chunking_strategy: nil,
+        # A description for the vector store. Can be used to describe the vector store's
+        # purpose.
+        description: nil,
         # The expiration policy for a vector store.
         expires_after: nil,
         # A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that
@@ -117,6 +129,7 @@ module OpenAI
                 OpenAI::AutoFileChunkingStrategyParam,
                 OpenAI::StaticFileChunkingStrategyObjectParam
               ),
+            description: String,
             expires_after: OpenAI::VectorStoreCreateParams::ExpiresAfter,
             file_ids: T::Array[String],
             metadata: T.nilable(T::Hash[Symbol, String]),

data/rbi/openai/resources/audio/transcriptions.rbi

@@ -20,6 +20,8 @@ module OpenAI
                 )
               ),
             include: T::Array[OpenAI::Audio::TranscriptionInclude::OrSymbol],
+            known_speaker_names: T::Array[String],
+            known_speaker_references: T::Array[String],
             language: String,
             prompt: String,
             response_format: OpenAI::AudioResponseFormat::OrSymbol,
@@ -39,20 +41,33 @@ module OpenAI
           # flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
           file:,
           # ID of the model to use. The options are `gpt-4o-transcribe`,
-          # `gpt-4o-mini-transcribe`, and `whisper-1` (which is powered by our open source
-          # Whisper V2 model).
+          # `gpt-4o-mini-transcribe`, `whisper-1` (which is powered by our open source
+          # Whisper V2 model), and `gpt-4o-transcribe-diarize`.
           model:,
           # Controls how the audio is cut into chunks. When set to `"auto"`, the server
           # first normalizes loudness and then uses voice activity detection (VAD) to choose
           # boundaries. `server_vad` object can be provided to tweak VAD detection
           # parameters manually. If unset, the audio is transcribed as a single block.
+          # Required when using `gpt-4o-transcribe-diarize` for inputs longer than 30
+          # seconds.
           chunking_strategy: nil,
           # Additional information to include in the transcription response. `logprobs` will
           # return the log probabilities of the tokens in the response to understand the
           # model's confidence in the transcription. `logprobs` only works with
           # response_format set to `json` and only with the models `gpt-4o-transcribe` and
-          # `gpt-4o-mini-transcribe`.
+          # `gpt-4o-mini-transcribe`. This field is not supported when using
+          # `gpt-4o-transcribe-diarize`.
           include: nil,
+          # Optional list of speaker names that correspond to the audio samples provided in
+          # `known_speaker_references[]`. Each entry should be a short identifier (for
+          # example `customer` or `agent`). Up to 4 speakers are supported.
+          known_speaker_names: nil,
+          # Optional list of audio samples (as
+          # [data URLs](https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Data_URLs))
+          # that contain known speaker references matching `known_speaker_names[]`. Each
+          # sample must be between 2 and 10 seconds, and can use any of the same input audio
+          # formats supported by `file`.
+          known_speaker_references: nil,
           # 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.
@@ -60,11 +75,14 @@ module OpenAI
           # 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.
+          # should match the audio language. This field is not supported when using
+          # `gpt-4o-transcribe-diarize`.
           prompt: nil,
           # The format of the output, in one of these options: `json`, `text`, `srt`,
-          # `verbose_json`, or `vtt`. For `gpt-4o-transcribe` and `gpt-4o-mini-transcribe`,
-          # the only supported format is `json`.
+          # `verbose_json`, `vtt`, or `diarized_json`. For `gpt-4o-transcribe` and
+          # `gpt-4o-mini-transcribe`, the only supported format is `json`. For
+          # `gpt-4o-transcribe-diarize`, the supported formats are `json`, `text`, and
+          # `diarized_json`, with `diarized_json` required to receive speaker annotations.
           response_format: nil,
           # The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
           # output more random, while lower values like 0.2 will make it more focused and
@@ -76,7 +94,8 @@ module OpenAI
           # `response_format` must be set `verbose_json` to use timestamp granularities.
           # Either or both of these options are supported: `word`, or `segment`. Note: There
           # is no additional latency for segment timestamps, but generating word timestamps
-          # incurs additional latency.
+          # incurs additional latency. This option is not available for
+          # `gpt-4o-transcribe-diarize`.
           timestamp_granularities: nil,
           # There is no need to provide `stream:`. Instead, use `#create_streaming` or
           # `#create` for streaming and non-streaming use cases, respectively.
@@ -101,6 +120,8 @@ module OpenAI
                 )
               ),
             include: T::Array[OpenAI::Audio::TranscriptionInclude::OrSymbol],
+            known_speaker_names: T::Array[String],
+            known_speaker_references: T::Array[String],
             language: String,
             prompt: String,
             response_format: OpenAI::AudioResponseFormat::OrSymbol,
@@ -122,20 +143,33 @@ module OpenAI
           # flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
           file:,
           # ID of the model to use. The options are `gpt-4o-transcribe`,
-          # `gpt-4o-mini-transcribe`, and `whisper-1` (which is powered by our open source
-          # Whisper V2 model).
+          # `gpt-4o-mini-transcribe`, `whisper-1` (which is powered by our open source
+          # Whisper V2 model), and `gpt-4o-transcribe-diarize`.
           model:,
           # Controls how the audio is cut into chunks. When set to `"auto"`, the server
           # first normalizes loudness and then uses voice activity detection (VAD) to choose
           # boundaries. `server_vad` object can be provided to tweak VAD detection
           # parameters manually. If unset, the audio is transcribed as a single block.
+          # Required when using `gpt-4o-transcribe-diarize` for inputs longer than 30
+          # seconds.
           chunking_strategy: nil,
           # Additional information to include in the transcription response. `logprobs` will
           # return the log probabilities of the tokens in the response to understand the
           # model's confidence in the transcription. `logprobs` only works with
           # response_format set to `json` and only with the models `gpt-4o-transcribe` and
-          # `gpt-4o-mini-transcribe`.
+          # `gpt-4o-mini-transcribe`. This field is not supported when using
+          # `gpt-4o-transcribe-diarize`.
           include: nil,
+          # Optional list of speaker names that correspond to the audio samples provided in
+          # `known_speaker_references[]`. Each entry should be a short identifier (for
+          # example `customer` or `agent`). Up to 4 speakers are supported.
+          known_speaker_names: nil,
+          # Optional list of audio samples (as
+          # [data URLs](https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Data_URLs))
+          # that contain known speaker references matching `known_speaker_names[]`. Each
+          # sample must be between 2 and 10 seconds, and can use any of the same input audio
+          # formats supported by `file`.
+          known_speaker_references: nil,
           # 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.
@@ -143,11 +177,14 @@ module OpenAI
           # 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.
+          # should match the audio language. This field is not supported when using
+          # `gpt-4o-transcribe-diarize`.
           prompt: nil,
           # The format of the output, in one of these options: `json`, `text`, `srt`,
-          # `verbose_json`, or `vtt`. For `gpt-4o-transcribe` and `gpt-4o-mini-transcribe`,
-          # the only supported format is `json`.
+          # `verbose_json`, `vtt`, or `diarized_json`. For `gpt-4o-transcribe` and
+          # `gpt-4o-mini-transcribe`, the only supported format is `json`. For
+          # `gpt-4o-transcribe-diarize`, the supported formats are `json`, `text`, and
+          # `diarized_json`, with `diarized_json` required to receive speaker annotations.
           response_format: nil,
           # The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
           # output more random, while lower values like 0.2 will make it more focused and
@@ -159,7 +196,8 @@ module OpenAI
           # `response_format` must be set `verbose_json` to use timestamp granularities.
           # Either or both of these options are supported: `word`, or `segment`. Note: There
           # is no additional latency for segment timestamps, but generating word timestamps
-          # incurs additional latency.
+          # incurs additional latency. This option is not available for
+          # `gpt-4o-transcribe-diarize`.
           timestamp_granularities: nil,
           # There is no need to provide `stream:`. Instead, use `#create_streaming` or
           # `#create` for streaming and non-streaming use cases, respectively.

data/rbi/openai/resources/vector_stores.rbi

@@ -17,6 +17,7 @@ module OpenAI
               OpenAI::AutoFileChunkingStrategyParam::OrHash,
               OpenAI::StaticFileChunkingStrategyObjectParam::OrHash
             ),
+          description: String,
           expires_after: OpenAI::VectorStoreCreateParams::ExpiresAfter::OrHash,
           file_ids: T::Array[String],
           metadata: T.nilable(T::Hash[Symbol, String]),
@@ -28,6 +29,9 @@ module OpenAI
         # The chunking strategy used to chunk the file(s). If not set, will use the `auto`
         # strategy. Only applicable if `file_ids` is non-empty.
         chunking_strategy: nil,
+        # A description for the vector store. Can be used to describe the vector store's
+        # purpose.
+        description: nil,
         # The expiration policy for a vector store.
         expires_after: nil,
         # A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that

data/sig/openai/models/audio/transcription_create_params.rbs

@@ -7,6 +7,8 @@ module OpenAI
           model: OpenAI::Models::Audio::TranscriptionCreateParams::model,
           chunking_strategy: OpenAI::Models::Audio::TranscriptionCreateParams::chunking_strategy?,
           include: ::Array[OpenAI::Models::Audio::transcription_include],
+          known_speaker_names: ::Array[String],
+          known_speaker_references: ::Array[String],
           language: String,
           prompt: String,
           response_format: OpenAI::Models::audio_response_format,
@@ -31,6 +33,14 @@ module OpenAI
           ::Array[OpenAI::Models::Audio::transcription_include]
         ) -> ::Array[OpenAI::Models::Audio::transcription_include]
 
+        attr_reader known_speaker_names: ::Array[String]?
+
+        def known_speaker_names=: (::Array[String]) -> ::Array[String]
+
+        attr_reader known_speaker_references: ::Array[String]?
+
+        def known_speaker_references=: (::Array[String]) -> ::Array[String]
+
         attr_reader language: String?
 
         def language=: (String) -> String
@@ -60,6 +70,8 @@ module OpenAI
           model: OpenAI::Models::Audio::TranscriptionCreateParams::model,
           ?chunking_strategy: OpenAI::Models::Audio::TranscriptionCreateParams::chunking_strategy?,
           ?include: ::Array[OpenAI::Models::Audio::transcription_include],
+          ?known_speaker_names: ::Array[String],
+          ?known_speaker_references: ::Array[String],
           ?language: String,
           ?prompt: String,
           ?response_format: OpenAI::Models::audio_response_format,
@@ -73,6 +85,8 @@ module OpenAI
           model: OpenAI::Models::Audio::TranscriptionCreateParams::model,
           chunking_strategy: OpenAI::Models::Audio::TranscriptionCreateParams::chunking_strategy?,
           include: ::Array[OpenAI::Models::Audio::transcription_include],
+          known_speaker_names: ::Array[String],
+          known_speaker_references: ::Array[String],
           language: String,
           prompt: String,
           response_format: OpenAI::Models::audio_response_format,

data/sig/openai/models/audio/transcription_create_response.rbs

@@ -2,7 +2,9 @@ module OpenAI
   module Models
     module Audio
       type transcription_create_response =
-        OpenAI::Audio::Transcription | OpenAI::Audio::TranscriptionVerbose
+        OpenAI::Audio::Transcription
+        | OpenAI::Audio::TranscriptionDiarized
+        | OpenAI::Audio::TranscriptionVerbose
 
       module TranscriptionCreateResponse
         extend OpenAI::Internal::Type::Union