telnyx 5.113.0 → 5.115.0

data/rbi/telnyx/models/ai/transcription_settings_config.rbi

@@ -21,6 +21,14 @@ module Telnyx
         sig { params(eager_eot_threshold: Float).void }
         attr_writer :eager_eot_threshold
 
+        # Available only for soniox/stt-rt-v4. When true, Soniox emits end-of-utterance
+        # events at the cadence configured by `max_endpoint_delay_ms`.
+        sig { returns(T.nilable(T::Boolean)) }
+        attr_reader :enable_endpoint_detection
+
+        sig { params(enable_endpoint_detection: T::Boolean).void }
+        attr_writer :enable_endpoint_detection
+
         # Available only for assemblyai/universal-streaming. Confidence level required to
         # trigger an end of turn. Higher values require more certainty before ending a
         # turn.
@@ -46,6 +54,14 @@ module Telnyx
         sig { params(eot_timeout_ms: Integer).void }
         attr_writer :eot_timeout_ms
 
+        # Available only for soniox/stt-rt-v4. When true, Soniox streams interim
+        # (non-final) results in addition to finalized transcripts.
+        sig { returns(T.nilable(T::Boolean)) }
+        attr_reader :interim_results
+
+        sig { params(interim_results: T::Boolean).void }
+        attr_writer :interim_results
+
         # Available only for deepgram/nova-3 and deepgram/flux. A comma-separated list of
         # key terms to boost for recognition during transcription. Helps improve accuracy
         # for domain-specific terminology, proper nouns, or uncommon words. This field may
@@ -59,6 +75,15 @@ module Telnyx
         sig { params(keyterm: String).void }
         attr_writer :keyterm
 
+        # Available only for soniox/stt-rt-v4. Maximum silence (in milliseconds) before
+        # Soniox emits an end-of-utterance event. Only honored when
+        # `enable_endpoint_detection` is true.
+        sig { returns(T.nilable(Integer)) }
+        attr_reader :max_endpoint_delay_ms
+
+        sig { params(max_endpoint_delay_ms: Integer).void }
+        attr_writer :max_endpoint_delay_ms
+
         # Available only for assemblyai/universal-streaming. Maximum duration of silence
         # in milliseconds before forcing an end of turn.
         sig { returns(T.nilable(Integer)) }
@@ -91,10 +116,13 @@ module Telnyx
         sig do
           params(
             eager_eot_threshold: Float,
+            enable_endpoint_detection: T::Boolean,
             end_of_turn_confidence_threshold: Float,
             eot_threshold: Float,
             eot_timeout_ms: Integer,
+            interim_results: T::Boolean,
             keyterm: String,
+            max_endpoint_delay_ms: Integer,
             max_turn_silence: Integer,
             min_turn_silence: Integer,
             numerals: T::Boolean,
@@ -106,6 +134,9 @@ module Telnyx
           # detection. Must be lower than or equal to eot_threshold. Setting this equal to
           # eot_threshold effectively disables eager end of turn.
           eager_eot_threshold: nil,
+          # Available only for soniox/stt-rt-v4. When true, Soniox emits end-of-utterance
+          # events at the cadence configured by `max_endpoint_delay_ms`.
+          enable_endpoint_detection: nil,
           # Available only for assemblyai/universal-streaming. Confidence level required to
           # trigger an end of turn. Higher values require more certainty before ending a
           # turn.
@@ -116,6 +147,9 @@ module Telnyx
           # Available only for deepgram/flux. Maximum milliseconds of silence before forcing
           # an end of turn, regardless of confidence.
           eot_timeout_ms: nil,
+          # Available only for soniox/stt-rt-v4. When true, Soniox streams interim
+          # (non-final) results in addition to finalized transcripts.
+          interim_results: nil,
           # Available only for deepgram/nova-3 and deepgram/flux. A comma-separated list of
           # key terms to boost for recognition during transcription. Helps improve accuracy
           # for domain-specific terminology, proper nouns, or uncommon words. This field may
@@ -124,6 +158,10 @@ module Telnyx
           # using mustache syntax (e.g. `Telnyx,{{customer_name}},VoIP`). Variables are
           # resolved at call time before the value is sent to the speech-to-text engine.
           keyterm: nil,
+          # Available only for soniox/stt-rt-v4. Maximum silence (in milliseconds) before
+          # Soniox emits an end-of-utterance event. Only honored when
+          # `enable_endpoint_detection` is true.
+          max_endpoint_delay_ms: nil,
           # Available only for assemblyai/universal-streaming. Maximum duration of silence
           # in milliseconds before forcing an end of turn.
           max_turn_silence: nil,
@@ -140,10 +178,13 @@ module Telnyx
           override.returns(
             {
               eager_eot_threshold: Float,
+              enable_endpoint_detection: T::Boolean,
               end_of_turn_confidence_threshold: Float,
               eot_threshold: Float,
               eot_timeout_ms: Integer,
+              interim_results: T::Boolean,
               keyterm: String,
+              max_endpoint_delay_ms: Integer,
               max_turn_silence: Integer,
               min_turn_silence: Integer,
               numerals: T::Boolean,

data/rbi/telnyx/models/ai/voice_settings.rbi

@@ -266,16 +266,28 @@ module Telnyx
             end
             attr_accessor :value
 
+            # Volume level for the predefined background audio. Supports values from 0.1 to
+            # 1.0 in 0.1 increments.
+            sig { returns(T.nilable(Float)) }
+            attr_reader :volume
+
+            sig { params(volume: Float).void }
+            attr_writer :volume
+
             sig do
               params(
                 value:
                   Telnyx::AI::VoiceSettings::BackgroundAudio::PredefinedMedia::Value::OrSymbol,
+                volume: Float,
                 type: Symbol
               ).returns(T.attached_class)
             end
             def self.new(
               # The predefined media to use. `silence` disables background audio.
               value:,
+              # Volume level for the predefined background audio. Supports values from 0.1 to
+              # 1.0 in 0.1 increments.
+              volume: nil,
               # Select from predefined media options.
               type: :predefined_media
             )
@@ -286,7 +298,8 @@ module Telnyx
                 {
                   type: Symbol,
                   value:
-                    Telnyx::AI::VoiceSettings::BackgroundAudio::PredefinedMedia::Value::OrSymbol
+                    Telnyx::AI::VoiceSettings::BackgroundAudio::PredefinedMedia::Value::OrSymbol,
+                  volume: Float
                 }
               )
             end

data/rbi/telnyx/models/ai_create_response_params.rbi

@@ -12,21 +12,21 @@ module Telnyx
         end
 
       sig { returns(T::Hash[Symbol, T.anything]) }
-      attr_accessor :params
+      attr_accessor :body
 
       sig do
         params(
-          params: T::Hash[Symbol, T.anything],
+          body: T::Hash[Symbol, T.anything],
           request_options: Telnyx::RequestOptions::OrHash
         ).returns(T.attached_class)
       end
-      def self.new(params:, request_options: {})
+      def self.new(body:, request_options: {})
       end
 
       sig do
         override.returns(
           {
-            params: T::Hash[Symbol, T.anything],
+            body: T::Hash[Symbol, T.anything],
             request_options: Telnyx::RequestOptions
           }
         )

data/rbi/telnyx/models/calls/transcription_config.rbi

@@ -14,7 +14,9 @@ module Telnyx
         # meaningful values depend on the selected transcription `model`. For
         # `deepgram/flux`, supported values are: `auto` (Telnyx language detection
         # controls the language hint), `multi` (no language hint), and language-specific
-        # hints `en`, `es`, `fr`, `de`, `hi`, `ru`, `pt`, `ja`, `it`, and `nl`.
+        # hints `en`, `es`, `fr`, `de`, `hi`, `ru`, `pt`, `ja`, `it`, and `nl`. For
+        # `soniox/stt-rt-v4`, `auto` omits the language hint and lets Soniox auto-detect;
+        # ISO 639-1 codes (e.g. `en`, `es`) bias detection toward that language.
         sig { returns(T.nilable(String)) }
         attr_reader :language
 
@@ -30,6 +32,8 @@ module Telnyx
         #   transcription.
         # - `assemblyai/universal-streaming` for live streaming transcription.
         # - `xai/grok-stt` for live streaming transcription.
+        # - `soniox/stt-rt-v4` for live streaming multilingual transcription with
+        #   automatic language detection.
         # - `azure/fast` and `azure/realtime`; Azure models require `region`, and
         #   unsupported regions require `api_key_ref`.
         # - `google/latest_long` for non-streaming multilingual transcription.
@@ -67,7 +71,9 @@ module Telnyx
           # meaningful values depend on the selected transcription `model`. For
           # `deepgram/flux`, supported values are: `auto` (Telnyx language detection
           # controls the language hint), `multi` (no language hint), and language-specific
-          # hints `en`, `es`, `fr`, `de`, `hi`, `ru`, `pt`, `ja`, `it`, and `nl`.
+          # hints `en`, `es`, `fr`, `de`, `hi`, `ru`, `pt`, `ja`, `it`, and `nl`. For
+          # `soniox/stt-rt-v4`, `auto` omits the language hint and lets Soniox auto-detect;
+          # ISO 639-1 codes (e.g. `en`, `es`) bias detection toward that language.
           language: nil,
           # The speech to text model to be used by the voice assistant. Supported models
           # include:
@@ -78,6 +84,8 @@ module Telnyx
           #   transcription.
           # - `assemblyai/universal-streaming` for live streaming transcription.
           # - `xai/grok-stt` for live streaming transcription.
+          # - `soniox/stt-rt-v4` for live streaming multilingual transcription with
+          #   automatic language detection.
           # - `azure/fast` and `azure/realtime`; Azure models require `region`, and
           #   unsupported regions require `api_key_ref`.
           # - `google/latest_long` for non-streaming multilingual transcription.
@@ -109,6 +117,8 @@ module Telnyx
         #   transcription.
         # - `assemblyai/universal-streaming` for live streaming transcription.
         # - `xai/grok-stt` for live streaming transcription.
+        # - `soniox/stt-rt-v4` for live streaming multilingual transcription with
+        #   automatic language detection.
         # - `azure/fast` and `azure/realtime`; Azure models require `region`, and
         #   unsupported regions require `api_key_ref`.
         # - `google/latest_long` for non-streaming multilingual transcription.
@@ -165,6 +175,11 @@ module Telnyx
               :"xai/grok-stt",
               Telnyx::Calls::TranscriptionConfig::Model::TaggedSymbol
             )
+          SONIOX_STT_RT_V4 =
+            T.let(
+              :"soniox/stt-rt-v4",
+              Telnyx::Calls::TranscriptionConfig::Model::TaggedSymbol
+            )
           AZURE_FAST =
             T.let(
               :"azure/fast",