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

# Process Pipeline

> All-in-one audio processing pipeline. Select steps via boolean flags. Returns a download URL for the processed audio.



## OpenAPI

````yaml POST /v1/enhancer/process/pipeline
openapi: 3.0.1
info:
  title: FineVoice API
  description: >-
    This is the documentation for the FineVoice API. If you have any questions
    or need assistance, please contact support@finevoice.ai.
  version: v1
servers:
  - url: https://apis.finevoice.ai
    description: FineVoice API Gateway
security:
  - BearerAuth: []
paths:
  /v1/enhancer/process/pipeline:
    post:
      tags:
        - Audio Enhancer
      summary: Process Pipeline
      description: >-
        All-in-one audio processing pipeline. Select steps via boolean flags.
        Execution order: (1) Speech Enhancement, (2) Remove Mouth Sounds, (3)
        Remove Long Silences, (4) Super Resolution, (5) Filler Words Removal,
        (6) Stuttering Removal, (7) Audio Normalization. Returns a download URL.
      operationId: ProcessPipeline
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/PipelineRequest'
      responses:
        '200':
          description: Processed audio download URL returned.
          content:
            application/json:
              schema:
                type: object
        '401':
          description: Unauthorized — invalid or missing API key.
        '422':
          description: Validation error — check request body fields.
components:
  schemas:
    PipelineRequest:
      type: object
      description: The all-in-one audio processing pipeline request payload.
      properties:
        url:
          type: string
          description: Audio file URL (http/https).
          example: https://example.com/audio.mp3
        step_speech_enhancement:
          type: boolean
          description: Enable Speech Enhancement step (noise reduction).
          default: true
        step_remove_mouth_sounds:
          type: boolean
          description: Enable Remove Mouth Sounds step.
          default: false
        step_remove_long_silences:
          type: boolean
          description: Enable Remove Long Silences step.
          default: false
        step_super_resolution:
          type: boolean
          description: Enable Super Resolution step (8 kHz to 48 kHz).
          default: false
        step_filler_words_remove:
          type: boolean
          description: Enable Filler Words Removal step.
          default: false
        step_stuttering_remove:
          type: boolean
          description: Enable Stuttering Removal step.
          default: false
        step_audio_normalization:
          type: boolean
          description: Enable Audio Normalization step.
          default: false
        enhancement_model:
          type: string
          description: >-
            Speech enhancement model: `MossFormer2_SE_48K`, `FRCRN_SE_16K`, or
            `MossFormerGAN_SE_16K`.
          default: FRCRN_SE_16K
        enhancement_use_vad:
          type: boolean
          description: Enable VAD preprocessing for speech enhancement.
          default: false
        silence_threshold_db:
          type: number
          description: Silence detection threshold in dBFS.
          default: -42
        long_silence_ms:
          type: integer
          description: Silences longer than this value (ms) are trimmed.
          default: 1200
        keep_silence_ms:
          type: integer
          description: Silence padding to preserve at boundaries (ms).
          default: 180
        filler_use_whisper:
          type: boolean
          description: Use Whisper for filler word detection.
          default: false
        filler_language:
          type: string
          description: Language code for filler word detection (e.g. `en`, `zh`).
          default: en
        filler_words_list:
          type: string
          description: Comma-separated filler words to remove. Empty = built-in defaults.
          default: um,uh,er,hmm,hm,ah,eh,mhm
        stutter_min_similarity:
          type: number
          minimum: 0.5
          maximum: 1
          description: Minimum similarity for stutter detection.
          default: 0.75
        norm_method:
          type: string
          description: 'Normalization method: `peak`, `rms`, or `both`.'
          default: peak
        norm_peak_db:
          type: number
          description: Target peak level in dBFS.
          default: -1
        norm_rms_db:
          type: number
          description: Target RMS level in dBFS.
          default: -20
        output_format:
          type: string
          description: 'Output format: `wav`, `mp3`, `flac`, or `m4a`.'
          default: wav
  securitySchemes:
    BearerAuth:
      type: http
      scheme: bearer
      bearerFormat: JWT
      description: 'Bearer token (API key). Format: `Bearer {your_api_key}`'

````