API docs / Audio models

speech-enhancer API

Removes noise and enhances speech quality from audio recordings. Offers three different models: MossFormer2_SE_48K for high-quality speech enhancement, CMGAN for noise reduction, and Resemble Denoiser with batch processing capabilities for throughput optimization. Supports various audio formats including wav, mp3, and m4a.

Model id
playmore/speech-enhancer
Provider
playmore
Output
Audio
Price
Depends on the options; the same as on voor.ai.

Request

POST https://voor.ai/v1/generations with "model": "playmore/speech-enhancer". Poll GET /v1/tasks/{id} or receive the signed webhook. See the overview for keys, idempotency, webhooks and errors.

curl https://voor.ai/v1/generations \
  -H "Authorization: Bearer $VOOR_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
  "model": "playmore/speech-enhancer",
  "input": {
    "reference_audios": [
      "https://example.com/input.png"
    ]
  },
  "callback_url": "https://example.com/webhooks/voor"
}'

# Then poll the task (or wait for the webhook):
curl https://voor.ai/v1/tasks/TASK_ID \
  -H "Authorization: Bearer $VOOR_API_KEY"

Input parameters

NameTypeDetails
reference_audiosrequiredarray of URLs

Noisy audio file URL, such as wav, mp3, or m4a.

Up to 1 files · Accepts audio/*

batch_process_chunksboolean

Only applies to resemble_denoiser; process 30-second chunks in batches for throughput.

Default: false

force_sliding_windowboolean

Only applies to mossformer2; otherwise ignored. Forces sliding-window decoding regardless of clip length.

Default: false

modelstring

Choose a model to run.

Default: mossformer2_se_48k

One of: mossformer2_se_48k, cmgan, resemble_denoiser