> ## Documentation Index
> Fetch the complete documentation index at: https://doc.featherhq.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Create a reusable named voice config



## OpenAPI

````yaml /api-reference/openapi.json post /v1/voice/saved-configs
openapi: 3.1.0
info:
  title: Feather API
  description: >-
    Unified customer experience platform API. Manages identity, conversations,
    memory, agents, procedures, policies, model routing, knowledge bases,
    integrations, and runtime execution.
  version: 1.87.0
servers:
  - url: https://api-sandbox.featherhq.com
    description: Sandbox
  - url: http://localhost:8000
    description: Local dev
security: []
paths:
  /v1/voice/saved-configs:
    post:
      tags:
        - voice
      summary: Create a reusable named voice config
      operationId: createSavedVoiceConfig
      requestBody:
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/SavedVoiceConfigCreate'
        required: true
      responses:
        '201':
          description: Successful Response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/VoicePipelineConfigResponse'
        '400':
          description: Bad request
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
        '401':
          description: Authentication required
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
        '422':
          description: Validation Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/HTTPValidationError'
      security:
        - APIKeyHeader: []
components:
  schemas:
    SavedVoiceConfigCreate:
      properties:
        voice_tuning:
          anyOf:
            - additionalProperties: true
              type: object
            - type: 'null'
          title: Voice Tuning
        stt_provider:
          anyOf:
            - type: string
              maxLength: 32
            - type: 'null'
          title: Stt Provider
        stt_model:
          anyOf:
            - type: string
              maxLength: 64
            - type: 'null'
          title: Stt Model
        stt_provider_data:
          anyOf:
            - additionalProperties: true
              type: object
            - type: 'null'
          title: Stt Provider Data
        tts_speed:
          anyOf:
            - type: number
              exclusiveMaximum: 10
              minimum: 0
            - type: 'null'
          title: Tts Speed
        tts_provider_data:
          anyOf:
            - $ref: '#/components/schemas/VoiceTTSProviderDataInput'
            - type: 'null'
        vad_provider_data:
          anyOf:
            - additionalProperties: true
              type: object
            - type: 'null'
          title: Vad Provider Data
        local_vad_params:
          anyOf:
            - $ref: '#/components/schemas/LocalVADParams'
            - type: 'null'
        turn_analyzer_params:
          anyOf:
            - $ref: '#/components/schemas/TurnAnalyzerParams'
            - type: 'null'
        pronunciation_overrides:
          anyOf:
            - additionalProperties:
                type: string
              type: object
            - type: 'null'
          title: Pronunciation Overrides
        noise_cancellation_enabled:
          anyOf:
            - type: boolean
            - type: 'null'
          title: Noise Cancellation Enabled
        noise_cancellation_provider:
          anyOf:
            - $ref: '#/components/schemas/NoiseCancellationProvider'
            - type: 'null'
        language:
          anyOf:
            - type: string
              maxLength: 16
            - type: 'null'
          title: Language
        locale:
          anyOf:
            - type: string
              maxLength: 16
            - type: 'null'
          title: Locale
        multilingual_enabled:
          anyOf:
            - type: boolean
            - type: 'null'
          title: Multilingual Enabled
        multilingual_languages:
          anyOf:
            - items:
                type: string
              type: array
              maxItems: 50
            - type: 'null'
          title: Multilingual Languages
        first_speaker:
          anyOf:
            - type: string
              enum:
                - agent
                - user
            - type: 'null'
          title: First Speaker
        mode:
          anyOf:
            - type: string
              enum:
                - static
                - dynamic
            - type: 'null'
          title: Mode
        text:
          anyOf:
            - type: string
              maxLength: 1000
            - type: 'null'
          title: Text
        instructions:
          anyOf:
            - type: string
              maxLength: 2000
            - type: 'null'
          title: Instructions
        interruptible:
          anyOf:
            - type: boolean
            - type: 'null'
          title: Interruptible
        ai_disclosure_text:
          anyOf:
            - type: string
              maxLength: 500
            - type: 'null'
          title: Ai Disclosure Text
        name:
          type: string
          maxLength: 255
          minLength: 1
          title: Name
        agent_id:
          type: 'null'
          title: Agent Id
        team_id:
          type: 'null'
          title: Team Id
        persona_id:
          type: 'null'
          title: Persona Id
        voice_id:
          type: string
          format: uuid
          title: Voice Id
      additionalProperties: false
      type: object
      required:
        - name
        - voice_id
      title: SavedVoiceConfigCreate
      description: |-
        Create a reusable, named (saved) voice config — no parent binding.

        Saved configs are owned by the org and can be referenced by multiple
        agents via ``Agent.voice_config_id``. They must have a non-empty name
        and must NOT set agent_id / team_id / persona_id (those are for inline
        configs only).
    VoicePipelineConfigResponse:
      properties:
        id:
          type: string
          format: uuid
          title: Id
        organization_id:
          anyOf:
            - type: string
              format: uuid
            - type: 'null'
          title: Organization Id
        name:
          anyOf:
            - type: string
            - type: 'null'
          title: Name
        agent_id:
          anyOf:
            - type: string
              format: uuid
            - type: 'null'
          title: Agent Id
        team_id:
          anyOf:
            - type: string
              format: uuid
            - type: 'null'
          title: Team Id
        persona_id:
          anyOf:
            - type: string
              format: uuid
            - type: 'null'
          title: Persona Id
        voice_id:
          type: string
          format: uuid
          title: Voice Id
        voice:
          $ref: '#/components/schemas/VoiceOut'
        voice_tuning:
          additionalProperties: true
          type: object
          title: Voice Tuning
        stt_provider:
          type: string
          title: Stt Provider
        stt_model:
          anyOf:
            - type: string
            - type: 'null'
          title: Stt Model
        stt_language:
          type: string
          title: Stt Language
        stt_provider_data:
          additionalProperties: true
          type: object
          title: Stt Provider Data
        tts_speed:
          anyOf:
            - type: number
            - type: 'null'
          title: Tts Speed
        tts_provider_data:
          additionalProperties: true
          type: object
          title: Tts Provider Data
        vad_provider:
          type: string
          title: Vad Provider
        vad_provider_data:
          additionalProperties: true
          type: object
          title: Vad Provider Data
        local_vad_params:
          additionalProperties: true
          type: object
          title: Local Vad Params
        turn_analyzer_params:
          additionalProperties: true
          type: object
          title: Turn Analyzer Params
        interruption_sensitivity:
          type: number
          title: Interruption Sensitivity
        hard_turn_timeout_ms:
          anyOf:
            - type: integer
            - type: 'null'
          title: Hard Turn Timeout Ms
        min_endpoint_delay_ms:
          type: integer
          title: Min Endpoint Delay Ms
        max_endpoint_delay_ms:
          type: integer
          title: Max Endpoint Delay Ms
        preemptive_generation_enabled:
          type: boolean
          title: Preemptive Generation Enabled
        end_call_phrases:
          items:
            type: string
          type: array
          title: End Call Phrases
        pronunciation_overrides:
          additionalProperties:
            type: string
          type: object
          title: Pronunciation Overrides
        noise_cancellation_enabled:
          type: boolean
          title: Noise Cancellation Enabled
        noise_cancellation_provider:
          $ref: '#/components/schemas/NoiseCancellationProvider'
        normalize_for_speech:
          type: boolean
          title: Normalize For Speech
        audio_in_sample_rate:
          type: integer
          title: Audio In Sample Rate
        audio_out_sample_rate:
          type: integer
          title: Audio Out Sample Rate
        metrics_enabled:
          type: boolean
          title: Metrics Enabled
        usage_metrics_enabled:
          type: boolean
          title: Usage Metrics Enabled
        heartbeats_period_secs:
          anyOf:
            - type: integer
            - type: 'null'
          title: Heartbeats Period Secs
        idle_pipeline_timeout_ms:
          anyOf:
            - type: integer
            - type: 'null'
          title: Idle Pipeline Timeout Ms
        language:
          type: string
          title: Language
        locale:
          anyOf:
            - type: string
            - type: 'null'
          title: Locale
        multilingual_enabled:
          type: boolean
          title: Multilingual Enabled
        multilingual_languages:
          items:
            type: string
          type: array
          title: Multilingual Languages
        linked_agent_count:
          type: integer
          title: Linked Agent Count
          default: 0
        linked_agent_names:
          items:
            type: string
          type: array
          title: Linked Agent Names
        linked_team_count:
          type: integer
          title: Linked Team Count
          default: 0
        linked_team_names:
          items:
            type: string
          type: array
          title: Linked Team Names
        metadata_:
          additionalProperties: true
          type: object
          title: Metadata
        first_speaking_config:
          anyOf:
            - $ref: '#/components/schemas/FirstSpeakingConfig'
            - type: 'null'
        warnings:
          items:
            type: string
          type: array
          title: Warnings
        effective_stt_model:
          anyOf:
            - type: string
            - type: 'null'
          title: Effective Stt Model
        effective_stt_provider_data:
          anyOf:
            - additionalProperties: true
              type: object
            - type: 'null'
          title: Effective Stt Provider Data
        effective_tts_model:
          anyOf:
            - type: string
            - type: 'null'
          title: Effective Tts Model
        created_at:
          type: string
          format: date-time
          title: Created At
        updated_at:
          type: string
          format: date-time
          title: Updated At
      additionalProperties: false
      type: object
      required:
        - id
        - voice_id
        - voice
        - stt_provider
        - stt_language
        - vad_provider
        - interruption_sensitivity
        - min_endpoint_delay_ms
        - max_endpoint_delay_ms
        - preemptive_generation_enabled
        - noise_cancellation_enabled
        - noise_cancellation_provider
        - normalize_for_speech
        - audio_in_sample_rate
        - audio_out_sample_rate
        - metrics_enabled
        - usage_metrics_enabled
        - language
        - multilingual_enabled
        - created_at
        - updated_at
      title: VoicePipelineConfigResponse
    ErrorResponse:
      properties:
        error:
          type: string
          title: Error
        message:
          type: string
          title: Message
        retryable:
          anyOf:
            - type: boolean
            - type: 'null'
          title: Retryable
        retry_after:
          anyOf:
            - type: integer
            - type: 'null'
          title: Retry After
      type: object
      required:
        - error
        - message
      title: ErrorResponse
      description: Standard error response returned by all API error handlers.
    HTTPValidationError:
      properties:
        detail:
          items:
            $ref: '#/components/schemas/ValidationError'
          type: array
          title: Detail
      type: object
      title: HTTPValidationError
    VoiceTTSProviderDataInput:
      properties:
        provider:
          anyOf:
            - type: string
              enum:
                - elevenlabs
                - inworld
            - type: 'null'
          title: Provider
        model:
          anyOf:
            - type: string
              maxLength: 64
            - type: 'null'
          title: Model
        expressive_speech:
          type: boolean
          title: Expressive Speech
          description: >-
            Enable managed expressive steering. Valid only for an Inworld voice
            using the effective inworld-tts-2 model.
          default: false
        expressive_style:
          anyOf:
            - $ref: '#/components/schemas/ExpressiveStyle'
            - type: 'null'
          description: >-
            Delivery style for expressive steering (e.g. warm, professional,
            upbeat). Requires expressive_speech enabled; defaults to 'warm'.
            Also sets Inworld's default delivery_mode unless one is pinned.
      additionalProperties: false
      type: object
      title: VoiceTTSProviderDataInput
      description: >-
        OpenAPI-visible TTS overrides accepted by voice-config CRUD.


        The linked ``voice_id`` remains the authority for which provider member
        is

        valid. The service resolves that row and validates this payload again

        against the provider-specific discriminated union, including model
        support

        and the Inworld TTS-2-only expressive-speech constraint.
    LocalVADParams:
      properties:
        confidence:
          type: number
          maximum: 1
          minimum: 0
          title: Confidence
          default: 0.7
        start_secs:
          type: number
          maximum: 2
          minimum: 0
          title: Start Secs
          default: 0.2
        stop_secs:
          type: number
          maximum: 3
          minimum: 0
          title: Stop Secs
          default: 0.5
        min_volume:
          type: number
          maximum: 1
          minimum: 0
          title: Min Volume
          default: 0.6
      additionalProperties: false
      type: object
      title: LocalVADParams
      description: Silero VAD tuning. Always applied for Deepgram Nova; ignored otherwise.
    TurnAnalyzerParams:
      properties:
        stop_secs:
          type: number
          maximum: 8
          minimum: 0.5
          title: Stop Secs
          default: 2
        pre_speech_ms:
          type: number
          maximum: 2000
          minimum: 0
          title: Pre Speech Ms
          default: 500
        max_duration_secs:
          type: number
          maximum: 30
          minimum: 2
          title: Max Duration Secs
          default: 8
      additionalProperties: false
      type: object
      title: TurnAnalyzerParams
      description: >-
        Smart-turn-v3 tuning. Always applied for Deepgram Nova; ignored
        otherwise.
    NoiseCancellationProvider:
      type: string
      enum:
        - rnnoise
        - dtln
      title: NoiseCancellationProvider
      description: >-
        Noise-suppression engine wired by
        ``audio_filters.build_audio_in_filter``.


        Applies only when ``noise_cancellation_enabled``. String values MUST
        match

        the provider strings the factory dispatches on.
    VoiceOut:
      properties:
        id:
          type: string
          format: uuid
          title: Id
        provider:
          type: string
          enum:
            - elevenlabs
            - inworld
          title: Provider
        external_id:
          type: string
          title: External Id
        name:
          type: string
          title: Name
        description:
          anyOf:
            - type: string
            - type: 'null'
          title: Description
        preview_url:
          anyOf:
            - type: string
            - type: 'null'
          title: Preview Url
        language:
          anyOf:
            - type: string
            - type: 'null'
          title: Language
        accent:
          anyOf:
            - type: string
            - type: 'null'
          title: Accent
        category:
          anyOf:
            - type: string
            - type: 'null'
          title: Category
        labels:
          additionalProperties:
            type: string
          type: object
          title: Labels
        supported_models:
          items:
            type: string
          type: array
          title: Supported Models
        is_active:
          type: boolean
          title: Is Active
          default: true
        created_at:
          type: string
          format: date-time
          title: Created At
        updated_at:
          type: string
          format: date-time
          title: Updated At
      type: object
      required:
        - id
        - provider
        - external_id
        - name
        - created_at
        - updated_at
      title: VoiceOut
    FirstSpeakingConfig:
      properties:
        first_speaker:
          type: string
          enum:
            - agent
            - user
          title: First Speaker
          default: agent
        mode:
          type: string
          enum:
            - static
            - dynamic
          title: Mode
          default: static
        text:
          anyOf:
            - type: string
              maxLength: 1000
            - type: 'null'
          title: Text
        instructions:
          anyOf:
            - type: string
              maxLength: 2000
            - type: 'null'
          title: Instructions
        interruptible:
          type: boolean
          title: Interruptible
          default: true
        ai_disclosure_text:
          anyOf:
            - type: string
              maxLength: 500
            - type: 'null'
          title: Ai Disclosure Text
      type: object
      title: FirstSpeakingConfig
      description: |-
        Config for the AI's opening turn (no user input yet).

        Lives inside ``ChannelConfig`` — first-speaking is per-channel.

        Mode-agnostic fields:
        - ``first_speaker``: ``"agent"`` opens the call with the rendered
          message; ``"user"`` skips the greeting (caller speaks first).
          Non-voice channels treat the inbound user message as the
          ``"user"`` opener and ignore this field at runtime.

        Greeting source (only applies when ``first_speaker="agent"``):
        - ``static`` mode: ``text`` is required and emitted verbatim
          (templated via the shared ``{{var}}`` engine, ``src.shared.templating.render``).
        - ``dynamic`` mode: ``instructions`` are sent to the model router
          with the agent persona to render an opening line (instructions
          are templated). When omitted/empty, ``instructions`` falls back
          to ``DEFAULT_DYNAMIC_FIRST_SPEAKING_INSTRUCTIONS`` so operators
          who just want a generic friendly opener don't have to author copy.

        Voice-only fields (ignored on non-audio channels):
        - ``interruptible``: whether the caller can barge in mid-greeting.
        - ``ai_disclosure_text``: regulatory disclosure (e.g. CA SB 1001),
          played as an uninterruptible TTS frame BEFORE the greeting.
          Disclosure plays even when ``first_speaker="user"`` because the
          regulation requires the disclosure regardless of who speaks first.
    ValidationError:
      properties:
        loc:
          items:
            anyOf:
              - type: string
              - type: integer
          type: array
          title: Location
        msg:
          type: string
          title: Message
        type:
          type: string
          title: Error Type
        input:
          title: Input
        ctx:
          type: object
          title: Context
      type: object
      required:
        - loc
        - msg
        - type
      title: ValidationError
    ExpressiveStyle:
      type: string
      enum:
        - warm
        - professional
        - upbeat
        - calm
        - empathetic
        - confident
        - dynamic
      title: ExpressiveStyle
      description: >-
        Selectable delivery style for Inworld TTS-2 expressive steering.


        Each value maps to a tone block interpolated into the shared expressive

        contract
        (:func:`src.runtime.speech_output.build_inworld_tts2_expressive_directive`)

        and, in the voice layer, to a default Inworld ``delivery_mode``.
        ``WARM``

        reproduces the original single-directive behavior and is the default
        applied

        when a config enables expressive speech without picking a style.
  securitySchemes:
    APIKeyHeader:
      type: apiKey
      in: header
      name: x-api-key

````