NVIDIA NIM · Capability

NVIDIA NIM Speech — ASR

Riva ASR (speech-to-text) NIMs — Parakeet, Canary.

NVIDIA NIM Speech — ASR is a Naftiko capability published by NVIDIA NIM, one of 11 capabilities the APIs.io network indexes for this provider. It bundles 1 operation across the POST method rooted at /v1/audio/transcriptions.

The capability includes 1 state-changing operation. Lead operation: Transcribe Audio. Can be deployed as a REST endpoint, MCP tool, or Agent Skill via Naftiko.

Tagged areas include NVIDIA, NIM, Riva, ASR, and Speech.

Run with Naftiko NVIDIANIMRivaASRSpeech

What You Can Do

POST

Createtranscription — Transcribe Audio

/v1/audio/transcriptions

MCP Tools

nvidia-nim-transcribe

Transcribe Audio

Capability Spec

naftiko: 1.0.0-alpha2
info:
  label: NVIDIA NIM Speech — ASR
  description: 'Riva ASR (speech-to-text) NIMs — Parakeet, Canary.'
  tags:
  - NVIDIA
  - NIM
  - Riva
  - ASR
  - Speech
  created: '2026-05-25'
  modified: '2026-05-25'
binds:
- namespace: env
  keys:
    NVIDIA_API_KEY: NVIDIA_API_KEY
capability:
  consumes:
  - type: http
    namespace: speech-asr
    baseUri: https://integrate.api.nvidia.com
    resources:
    - name: transcriptions
      path: /v1/audio/transcriptions
      operations:
      - name: createtranscription
        method: POST
        description: Transcribe Audio
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: body
          in: body
          type: object
          required: true
    authentication:
      type: bearer
      value: '{{env.NVIDIA_API_KEY}}'
      placement: header
  exposes:
  - type: rest
    namespace: speech-asr-rest
    port: 8080
    resources:
    - path: /v1/audio/transcriptions
      name: transcriptions
      operations:
      - method: POST
        name: createtranscription
        description: Transcribe Audio
        call: speech-asr.createtranscription
        with:
          body: rest.body
        outputParameters:
        - type: object
          mapping: $.
  - type: mcp
    namespace: speech-asr-mcp
    port: 9090
    transport: http
    tools:
    - name: nvidia-nim-transcribe
      description: Transcribe Audio
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: speech-asr.createtranscription
      with:
        body: tools.body
      outputParameters:
      - type: object
        mapping: $.