telnyx 5.169.0 → 5.171.0

data/rbi/telnyx/models/ai/transcription_settings.rbi

@@ -17,6 +17,38 @@ module Telnyx
         sig { params(api_key_ref: String).void }
         attr_writer :api_key_ref
 
+        # A second speech-to-text model that transcribes alongside `transcription.model`,
+        # and the rule that decides which transcript the assistant uses.
+        sig do
+          returns(T.nilable(Telnyx::AI::TranscriptionSettings::Challenger))
+        end
+        attr_reader :challenger
+
+        sig do
+          params(
+            challenger:
+              T.nilable(Telnyx::AI::TranscriptionSettings::Challenger::OrHash)
+          ).void
+        end
+        attr_writer :challenger
+
+        # Up to 3 streaming models that take over transcription, in this order, when the
+        # model in use fails, at the start of a call or mid-call. `model` must be a
+        # streaming model too, and must support `language` alongside other models. On
+        # update, a list replaces the stored one: omit the field to keep the stored list,
+        # or send `null` or `[]` to remove it. When an update changes `model` or
+        # `language`, stored fallbacks that no longer fit are removed without an error.
+        # Can't be combined with `challenger`, the language booster; to replace a stored
+        # language booster, send `challenger: null` in the same request.
+        sig do
+          returns(
+            T.nilable(
+              T::Array[Telnyx::AI::TranscriptionSettings::FallbackModel]
+            )
+          )
+        end
+        attr_accessor :fallback_models
+
         # The language of the audio to be transcribed. If not set, or if set to `auto`,
         # supported models will automatically detect the language. For `deepgram/flux`,
         # supported values are: `auto` (Telnyx language detection controls the language
@@ -63,6 +95,7 @@ module Telnyx
         # - `reson8/turns` is a turn-based streaming model covering 10 European languages
         #   with automatic language detection.
         # - `cohere/ar-stt` is a non-streaming Arabic and English transcription model.
+        # - `telnyx/basira` is a non-streaming Arabic transcription model.
         sig do
           returns(T.nilable(Telnyx::AI::TranscriptionSettings::Model::OrSymbol))
         end
@@ -92,6 +125,14 @@ module Telnyx
         sig do
           params(
             api_key_ref: String,
+            challenger:
+              T.nilable(Telnyx::AI::TranscriptionSettings::Challenger::OrHash),
+            fallback_models:
+              T.nilable(
+                T::Array[
+                  Telnyx::AI::TranscriptionSettings::FallbackModel::OrHash
+                ]
+              ),
             language: String,
             model: Telnyx::AI::TranscriptionSettings::Model::OrSymbol,
             region: String,
@@ -102,6 +143,18 @@ module Telnyx
           # Integration secret identifier for the transcription provider API key. Currently
           # used for Azure transcription regions that require a customer-provided API key.
           api_key_ref: nil,
+          # A second speech-to-text model that transcribes alongside `transcription.model`,
+          # and the rule that decides which transcript the assistant uses.
+          challenger: nil,
+          # Up to 3 streaming models that take over transcription, in this order, when the
+          # model in use fails, at the start of a call or mid-call. `model` must be a
+          # streaming model too, and must support `language` alongside other models. On
+          # update, a list replaces the stored one: omit the field to keep the stored list,
+          # or send `null` or `[]` to remove it. When an update changes `model` or
+          # `language`, stored fallbacks that no longer fit are removed without an error.
+          # Can't be combined with `challenger`, the language booster; to replace a stored
+          # language booster, send `challenger: null` in the same request.
+          fallback_models: nil,
           # The language of the audio to be transcribed. If not set, or if set to `auto`,
           # supported models will automatically detect the language. For `deepgram/flux`,
           # supported values are: `auto` (Telnyx language detection controls the language
@@ -143,6 +196,7 @@ module Telnyx
           # - `reson8/turns` is a turn-based streaming model covering 10 European languages
           #   with automatic language detection.
           # - `cohere/ar-stt` is a non-streaming Arabic and English transcription model.
+          # - `telnyx/basira` is a non-streaming Arabic transcription model.
           model: nil,
           # Region on third party cloud providers (currently Azure) if using one of their
           # models. Some regions require `api_key_ref`.
@@ -155,6 +209,12 @@ module Telnyx
           override.returns(
             {
               api_key_ref: String,
+              challenger:
+                T.nilable(Telnyx::AI::TranscriptionSettings::Challenger),
+              fallback_models:
+                T.nilable(
+                  T::Array[Telnyx::AI::TranscriptionSettings::FallbackModel]
+                ),
               language: String,
               model: Telnyx::AI::TranscriptionSettings::Model::OrSymbol,
               region: String,
@@ -165,6 +225,504 @@ module Telnyx
         def to_hash
         end
 
+        class Challenger < Telnyx::Internal::Type::BaseModel
+          OrHash =
+            T.type_alias do
+              T.any(
+                Telnyx::AI::TranscriptionSettings::Challenger,
+                Telnyx::Internal::AnyHash
+              )
+            end
+
+          # The language booster's model. It must be the same kind of model as
+          # `transcription.model`: both streaming (`deepgram/flux`, `deepgram/nova-3`,
+          # `deepgram/nova-2`, `assemblyai/universal-3-5-pro` or its legacy alias
+          # `assemblyai/universal-streaming`, `xai/grok-stt`, `soniox/stt-rt-v4`,
+          # `soniox/stt-rt-v5`, `humain/realtime`, `reson8/turns`) or both non-streaming
+          # (`azure/fast`, `nvidia/parakeet-v3`, `omi-health/omi-med-stt-v1`,
+          # `cohere/ar-stt`, `distil-whisper/distil-large-v2`,
+          # `openai/whisper-large-v3-turbo`, `telnyx/basira`). It can be the same model as
+          # `transcription.model` on a different `language`.
+          sig do
+            returns(
+              Telnyx::AI::TranscriptionSettings::Challenger::Model::OrSymbol
+            )
+          end
+          attr_accessor :model
+
+          # The language this model transcribes. Omit it or set it to `null` to use the
+          # language of `transcription.model`. The request is rejected when this model
+          # doesn't support the language it would run. It is also rejected when it would run
+          # the same model on the same language as `transcription.model`.
+          sig { returns(T.nilable(String)) }
+          attr_accessor :language
+
+          # How the assistant picks the transcript it uses. The models are compared on how
+          # complete and confident their transcripts are, not on language, so the rules work
+          # best when both models understand the callers' language.
+          #
+          # - `best_turn` (default): both models transcribe the whole call. Each turn uses
+          #   the language booster's transcript only when it scores higher than the
+          #   transcript of `transcription.model` (clearly higher with non-streaming
+          #   models). With streaming models, `transcription.model` also decides when each
+          #   turn ends. Available for every pair.
+          # - `best_engine`: both models transcribe the first turns, then the call continues
+          #   alone on the model whose transcripts scored higher. If neither clearly leads,
+          #   `transcription.model` continues. Streaming models only.
+          # - `merge_words`: both models transcribe each utterance and their words are
+          #   merged, keeping Arabic and English spoken in the same sentence. Available only
+          #   for `telnyx/basira` with `cohere/ar-stt`, in either order. The pair runs on
+          #   the language that applies to `telnyx/basira` (its own, or that of
+          #   `transcription.model`), which must be Arabic (`ar` or an `ar-` locale),
+          #   `multi`, or `auto`.
+          sig do
+            returns(
+              T.nilable(
+                Telnyx::AI::TranscriptionSettings::Challenger::Rule::OrSymbol
+              )
+            )
+          end
+          attr_reader :rule
+
+          sig do
+            params(
+              rule:
+                Telnyx::AI::TranscriptionSettings::Challenger::Rule::OrSymbol
+            ).void
+          end
+          attr_writer :rule
+
+          # Settings for the language booster, with the same fields and limits as
+          # `transcription.settings`. Fields that don't apply to this model's provider are
+          # dropped, and the provider's defaults fill in the rest. Omit it or set it to
+          # `null` to use the settings of `transcription.model` where they apply to this
+          # model.
+          sig { returns(T.nilable(Telnyx::AI::TranscriptionSettingsConfig)) }
+          attr_reader :settings
+
+          sig do
+            params(
+              settings:
+                T.nilable(Telnyx::AI::TranscriptionSettingsConfig::OrHash)
+            ).void
+          end
+          attr_writer :settings
+
+          # A second speech-to-text model that transcribes alongside `transcription.model`,
+          # and the rule that decides which transcript the assistant uses.
+          sig do
+            params(
+              model:
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::OrSymbol,
+              language: T.nilable(String),
+              rule:
+                Telnyx::AI::TranscriptionSettings::Challenger::Rule::OrSymbol,
+              settings:
+                T.nilable(Telnyx::AI::TranscriptionSettingsConfig::OrHash)
+            ).returns(T.attached_class)
+          end
+          def self.new(
+            # The language booster's model. It must be the same kind of model as
+            # `transcription.model`: both streaming (`deepgram/flux`, `deepgram/nova-3`,
+            # `deepgram/nova-2`, `assemblyai/universal-3-5-pro` or its legacy alias
+            # `assemblyai/universal-streaming`, `xai/grok-stt`, `soniox/stt-rt-v4`,
+            # `soniox/stt-rt-v5`, `humain/realtime`, `reson8/turns`) or both non-streaming
+            # (`azure/fast`, `nvidia/parakeet-v3`, `omi-health/omi-med-stt-v1`,
+            # `cohere/ar-stt`, `distil-whisper/distil-large-v2`,
+            # `openai/whisper-large-v3-turbo`, `telnyx/basira`). It can be the same model as
+            # `transcription.model` on a different `language`.
+            model:,
+            # The language this model transcribes. Omit it or set it to `null` to use the
+            # language of `transcription.model`. The request is rejected when this model
+            # doesn't support the language it would run. It is also rejected when it would run
+            # the same model on the same language as `transcription.model`.
+            language: nil,
+            # How the assistant picks the transcript it uses. The models are compared on how
+            # complete and confident their transcripts are, not on language, so the rules work
+            # best when both models understand the callers' language.
+            #
+            # - `best_turn` (default): both models transcribe the whole call. Each turn uses
+            #   the language booster's transcript only when it scores higher than the
+            #   transcript of `transcription.model` (clearly higher with non-streaming
+            #   models). With streaming models, `transcription.model` also decides when each
+            #   turn ends. Available for every pair.
+            # - `best_engine`: both models transcribe the first turns, then the call continues
+            #   alone on the model whose transcripts scored higher. If neither clearly leads,
+            #   `transcription.model` continues. Streaming models only.
+            # - `merge_words`: both models transcribe each utterance and their words are
+            #   merged, keeping Arabic and English spoken in the same sentence. Available only
+            #   for `telnyx/basira` with `cohere/ar-stt`, in either order. The pair runs on
+            #   the language that applies to `telnyx/basira` (its own, or that of
+            #   `transcription.model`), which must be Arabic (`ar` or an `ar-` locale),
+            #   `multi`, or `auto`.
+            rule: nil,
+            # Settings for the language booster, with the same fields and limits as
+            # `transcription.settings`. Fields that don't apply to this model's provider are
+            # dropped, and the provider's defaults fill in the rest. Omit it or set it to
+            # `null` to use the settings of `transcription.model` where they apply to this
+            # model.
+            settings: nil
+          )
+          end
+
+          sig do
+            override.returns(
+              {
+                model:
+                  Telnyx::AI::TranscriptionSettings::Challenger::Model::OrSymbol,
+                language: T.nilable(String),
+                rule:
+                  Telnyx::AI::TranscriptionSettings::Challenger::Rule::OrSymbol,
+                settings: T.nilable(Telnyx::AI::TranscriptionSettingsConfig)
+              }
+            )
+          end
+          def to_hash
+          end
+
+          # The language booster's model. It must be the same kind of model as
+          # `transcription.model`: both streaming (`deepgram/flux`, `deepgram/nova-3`,
+          # `deepgram/nova-2`, `assemblyai/universal-3-5-pro` or its legacy alias
+          # `assemblyai/universal-streaming`, `xai/grok-stt`, `soniox/stt-rt-v4`,
+          # `soniox/stt-rt-v5`, `humain/realtime`, `reson8/turns`) or both non-streaming
+          # (`azure/fast`, `nvidia/parakeet-v3`, `omi-health/omi-med-stt-v1`,
+          # `cohere/ar-stt`, `distil-whisper/distil-large-v2`,
+          # `openai/whisper-large-v3-turbo`, `telnyx/basira`). It can be the same model as
+          # `transcription.model` on a different `language`.
+          module Model
+            extend Telnyx::Internal::Type::Enum
+
+            TaggedSymbol =
+              T.type_alias do
+                T.all(
+                  Symbol,
+                  Telnyx::AI::TranscriptionSettings::Challenger::Model
+                )
+              end
+            OrSymbol = T.type_alias { T.any(Symbol, String) }
+
+            DEEPGRAM_FLUX =
+              T.let(
+                :"deepgram/flux",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            DEEPGRAM_NOVA_3 =
+              T.let(
+                :"deepgram/nova-3",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            DEEPGRAM_NOVA_2 =
+              T.let(
+                :"deepgram/nova-2",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            AZURE_FAST =
+              T.let(
+                :"azure/fast",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            ASSEMBLYAI_UNIVERSAL_3_5_PRO =
+              T.let(
+                :"assemblyai/universal-3-5-pro",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            ASSEMBLYAI_UNIVERSAL_STREAMING =
+              T.let(
+                :"assemblyai/universal-streaming",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            XAI_GROK_STT =
+              T.let(
+                :"xai/grok-stt",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            SONIOX_STT_RT_V4 =
+              T.let(
+                :"soniox/stt-rt-v4",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            SONIOX_STT_RT_V5 =
+              T.let(
+                :"soniox/stt-rt-v5",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            NVIDIA_PARAKEET_V3 =
+              T.let(
+                :"nvidia/parakeet-v3",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            OMI_HEALTH_OMI_MED_STT_V1 =
+              T.let(
+                :"omi-health/omi-med-stt-v1",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            HUMAIN_REALTIME =
+              T.let(
+                :"humain/realtime",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            RESON8_TURNS =
+              T.let(
+                :"reson8/turns",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            COHERE_AR_STT =
+              T.let(
+                :"cohere/ar-stt",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            TELNYX_BASIRA =
+              T.let(
+                :"telnyx/basira",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            DISTIL_WHISPER_DISTIL_LARGE_V2 =
+              T.let(
+                :"distil-whisper/distil-large-v2",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+            OPENAI_WHISPER_LARGE_V3_TURBO =
+              T.let(
+                :"openai/whisper-large-v3-turbo",
+                Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+              )
+
+            sig do
+              override.returns(
+                T::Array[
+                  Telnyx::AI::TranscriptionSettings::Challenger::Model::TaggedSymbol
+                ]
+              )
+            end
+            def self.values
+            end
+          end
+
+          # How the assistant picks the transcript it uses. The models are compared on how
+          # complete and confident their transcripts are, not on language, so the rules work
+          # best when both models understand the callers' language.
+          #
+          # - `best_turn` (default): both models transcribe the whole call. Each turn uses
+          #   the language booster's transcript only when it scores higher than the
+          #   transcript of `transcription.model` (clearly higher with non-streaming
+          #   models). With streaming models, `transcription.model` also decides when each
+          #   turn ends. Available for every pair.
+          # - `best_engine`: both models transcribe the first turns, then the call continues
+          #   alone on the model whose transcripts scored higher. If neither clearly leads,
+          #   `transcription.model` continues. Streaming models only.
+          # - `merge_words`: both models transcribe each utterance and their words are
+          #   merged, keeping Arabic and English spoken in the same sentence. Available only
+          #   for `telnyx/basira` with `cohere/ar-stt`, in either order. The pair runs on
+          #   the language that applies to `telnyx/basira` (its own, or that of
+          #   `transcription.model`), which must be Arabic (`ar` or an `ar-` locale),
+          #   `multi`, or `auto`.
+          module Rule
+            extend Telnyx::Internal::Type::Enum
+
+            TaggedSymbol =
+              T.type_alias do
+                T.all(
+                  Symbol,
+                  Telnyx::AI::TranscriptionSettings::Challenger::Rule
+                )
+              end
+            OrSymbol = T.type_alias { T.any(Symbol, String) }
+
+            BEST_TURN =
+              T.let(
+                :best_turn,
+                Telnyx::AI::TranscriptionSettings::Challenger::Rule::TaggedSymbol
+              )
+            BEST_ENGINE =
+              T.let(
+                :best_engine,
+                Telnyx::AI::TranscriptionSettings::Challenger::Rule::TaggedSymbol
+              )
+            MERGE_WORDS =
+              T.let(
+                :merge_words,
+                Telnyx::AI::TranscriptionSettings::Challenger::Rule::TaggedSymbol
+              )
+
+            sig do
+              override.returns(
+                T::Array[
+                  Telnyx::AI::TranscriptionSettings::Challenger::Rule::TaggedSymbol
+                ]
+              )
+            end
+            def self.values
+            end
+          end
+        end
+
+        class FallbackModel < Telnyx::Internal::Type::BaseModel
+          OrHash =
+            T.type_alias do
+              T.any(
+                Telnyx::AI::TranscriptionSettings::FallbackModel,
+                Telnyx::Internal::AnyHash
+              )
+            end
+
+          # The fallback model. It must be a streaming model other than
+          # `transcription.model` and the other fallbacks: `deepgram/flux`,
+          # `deepgram/nova-3`, `deepgram/nova-2`, `assemblyai/universal-3-5-pro` (or its
+          # legacy alias `assemblyai/universal-streaming`), `xai/grok-stt`,
+          # `soniox/stt-rt-v4`, `soniox/stt-rt-v5`, `humain/realtime`, or `reson8/turns`.
+          sig do
+            returns(
+              Telnyx::AI::TranscriptionSettings::FallbackModel::Model::OrSymbol
+            )
+          end
+          attr_accessor :model
+
+          # The language the fallback transcribes. Omit it or set it to `null` to use the
+          # language of `transcription.model`. The request is rejected when the fallback
+          # model doesn't support the language it would run.
+          sig { returns(T.nilable(String)) }
+          attr_accessor :language
+
+          # Settings for the fallback, with the same fields and limits as
+          # `transcription.settings`. Fields that don't apply to this model's provider are
+          # dropped, and the provider's defaults fill in the rest. Omit it or set it to
+          # `null` to use the settings of `transcription.model` where they apply to this
+          # model.
+          sig { returns(T.nilable(Telnyx::AI::TranscriptionSettingsConfig)) }
+          attr_reader :settings
+
+          sig do
+            params(
+              settings:
+                T.nilable(Telnyx::AI::TranscriptionSettingsConfig::OrHash)
+            ).void
+          end
+          attr_writer :settings
+
+          # A streaming speech-to-text model that takes over transcription when the model in
+          # use fails.
+          sig do
+            params(
+              model:
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::OrSymbol,
+              language: T.nilable(String),
+              settings:
+                T.nilable(Telnyx::AI::TranscriptionSettingsConfig::OrHash)
+            ).returns(T.attached_class)
+          end
+          def self.new(
+            # The fallback model. It must be a streaming model other than
+            # `transcription.model` and the other fallbacks: `deepgram/flux`,
+            # `deepgram/nova-3`, `deepgram/nova-2`, `assemblyai/universal-3-5-pro` (or its
+            # legacy alias `assemblyai/universal-streaming`), `xai/grok-stt`,
+            # `soniox/stt-rt-v4`, `soniox/stt-rt-v5`, `humain/realtime`, or `reson8/turns`.
+            model:,
+            # The language the fallback transcribes. Omit it or set it to `null` to use the
+            # language of `transcription.model`. The request is rejected when the fallback
+            # model doesn't support the language it would run.
+            language: nil,
+            # Settings for the fallback, with the same fields and limits as
+            # `transcription.settings`. Fields that don't apply to this model's provider are
+            # dropped, and the provider's defaults fill in the rest. Omit it or set it to
+            # `null` to use the settings of `transcription.model` where they apply to this
+            # model.
+            settings: nil
+          )
+          end
+
+          sig do
+            override.returns(
+              {
+                model:
+                  Telnyx::AI::TranscriptionSettings::FallbackModel::Model::OrSymbol,
+                language: T.nilable(String),
+                settings: T.nilable(Telnyx::AI::TranscriptionSettingsConfig)
+              }
+            )
+          end
+          def to_hash
+          end
+
+          # The fallback model. It must be a streaming model other than
+          # `transcription.model` and the other fallbacks: `deepgram/flux`,
+          # `deepgram/nova-3`, `deepgram/nova-2`, `assemblyai/universal-3-5-pro` (or its
+          # legacy alias `assemblyai/universal-streaming`), `xai/grok-stt`,
+          # `soniox/stt-rt-v4`, `soniox/stt-rt-v5`, `humain/realtime`, or `reson8/turns`.
+          module Model
+            extend Telnyx::Internal::Type::Enum
+
+            TaggedSymbol =
+              T.type_alias do
+                T.all(
+                  Symbol,
+                  Telnyx::AI::TranscriptionSettings::FallbackModel::Model
+                )
+              end
+            OrSymbol = T.type_alias { T.any(Symbol, String) }
+
+            DEEPGRAM_FLUX =
+              T.let(
+                :"deepgram/flux",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            DEEPGRAM_NOVA_3 =
+              T.let(
+                :"deepgram/nova-3",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            DEEPGRAM_NOVA_2 =
+              T.let(
+                :"deepgram/nova-2",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            ASSEMBLYAI_UNIVERSAL_3_5_PRO =
+              T.let(
+                :"assemblyai/universal-3-5-pro",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            ASSEMBLYAI_UNIVERSAL_STREAMING =
+              T.let(
+                :"assemblyai/universal-streaming",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            XAI_GROK_STT =
+              T.let(
+                :"xai/grok-stt",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            SONIOX_STT_RT_V4 =
+              T.let(
+                :"soniox/stt-rt-v4",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            SONIOX_STT_RT_V5 =
+              T.let(
+                :"soniox/stt-rt-v5",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            HUMAIN_REALTIME =
+              T.let(
+                :"humain/realtime",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+            RESON8_TURNS =
+              T.let(
+                :"reson8/turns",
+                Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+              )
+
+            sig do
+              override.returns(
+                T::Array[
+                  Telnyx::AI::TranscriptionSettings::FallbackModel::Model::TaggedSymbol
+                ]
+              )
+            end
+            def self.values
+            end
+          end
+        end
+
         # The speech to text model to be used by the voice assistant. All Deepgram models
         # are run on-premise.
         #
@@ -188,6 +746,7 @@ module Telnyx
         # - `reson8/turns` is a turn-based streaming model covering 10 European languages
         #   with automatic language detection.
         # - `cohere/ar-stt` is a non-streaming Arabic and English transcription model.
+        # - `telnyx/basira` is a non-streaming Arabic transcription model.
         module Model
           extend Telnyx::Internal::Type::Enum
 
@@ -267,6 +826,11 @@ module Telnyx
               :"cohere/ar-stt",
               Telnyx::AI::TranscriptionSettings::Model::TaggedSymbol
             )
+          TELNYX_BASIRA =
+            T.let(
+              :"telnyx/basira",
+              Telnyx::AI::TranscriptionSettings::Model::TaggedSymbol
+            )
           DISTIL_WHISPER_DISTIL_LARGE_V2 =
             T.let(
               :"distil-whisper/distil-large-v2",