Get started
with AudicLabs

Speech AI you can call from anywhere: clean audio today, with transcription, synthesis and speaker intelligence on the way. One gateway, one key.

Create API key Quickstart

curl https://internal.audiclabs.com/v1/noise-suppressor/denoise \
  -H "Authorization: Bearer $AUDICLABS_API_KEY" \
  -F file=@noisy.wav \
  -F model=audic-2 \
  -o clean.wav

APIs

Noise Suppressor

Available

noise-suppressor

Remove background noise from speech, one file or a batch of 25, at any sample rate.

Models: Audic 1 to 4, Audic 2 HD
Sample rates: 16 kHz and 48 kHz
Formats in: WAV, FLAC, OGG, MP3, M4A, WebM
Per request: 1 file or up to 25

Read docs Try in console

API gateway

/v1/{service}/{endpoint}

Every model is published behind one host and one access token, with scopes, limits and logs.

Auth: Bearer token or X-API-Key
Scopes: Per service or endpoint
Limits: Per-minute rate, monthly quota
Tracing: X-Request-Id on every call

Read docs Open API explorer

Coming to your workspace

Enabled per workspace. Ask your admin for access.

Speech to Text · Hindi Speech to Text · English Speech to Text · Non-Indic Multilingual Speech to Text · Indic Multilingual Voice Activity Detection Speaker Diarization Text to Speech · English Text to Speech · Hindi Text to Speech · Non-Indic Multilingual Text to Speech · Indic Multilingual Speaker Recognition

Next steps

Quickstart

Get a key and clean your first file in three steps.

Remove noise (bulk)

Clean up to 25 files in one request.

Authentication

Tokens, scopes and the headers to send.

Errors

One error format, and what each type means.