VOICE & TRANSCRIPTION

Build voice products that listen, understand and respond.

Connect accurate transcription and natural speech output through one platform identity—ideal for meetings, media, support and conversational interfaces.

Start building ↗Open the Audio API reference
PRODUCTION ENDPOINTPOST /v1/audio
STTTRANSCRIPTION
TTSSPEECH OUTPUT
1 KEYVOICE STACK
Production advantage

Move faster without
shrinking your options.

AIx keeps identity, integration and measured billing consistent while your workload uses the model best suited to the job.

01

Speech in both directions

Transcribe uploaded audio and generate natural voice output with modality-specific endpoints.

02

One operational surface

Keep voice, chat and reasoning usage inside the same keys, limits and billing workflow.

03

Composable voice architecture

Combine transcription, a chat model and speech generation for complete conversational products.

Recommended models

Start with a model
matched to the work.

Use these as practical starting points, then evaluate quality, latency and unit economics against your own traffic.

AVAILABLE ON AIX

Whisper Large V3

Accurate multilingual transcription

Model API details →AVAILABLE ON AIX

Kokoro 82M

Natural, efficient speech generation

Model API details →AVAILABLE ON AIX

Gemini 2.5 Flash

Fast reasoning between voice turns

Model API details →
Reference architecture

From request
to production.

  1. 01
    Capture and normalize audio

    Record a supported format, remove unusable silence and preserve the original file for auditability.

  2. 02
    Transcribe and reason

    Convert speech to text, then send the transcript to the right chat model with conversation context.

  3. 03
    Generate the spoken response

    Transform the final answer into speech and stream or deliver it through your application.

QUICKSTARTPOST /v1/audio
curl https://api.aix.theaimart.co/v1/audio/transcriptions \
  -H "Authorization: Bearer $AIX_KEY" \
  -F file=@meeting.mp3 \
  -F model=openai/whisper-large-v3
Open the Audio API reference →
FAQ

Questions, answered.

Does AIx support both speech-to-text and text-to-speech?+

Yes. AIx exposes dedicated transcription and speech generation endpoints for supported models.

How is transcription priced?+

Transcription models are commonly metered by audio minute. Confirm the live price and unit in the AIx catalog.

Can voice workflows call a chat model too?+

Yes. A complete voice assistant can transcribe audio, call a chat model and generate speech with the same AIx key.

Ship the first request today.

One key. One API contract. The AIx model network behind it.

Get your API key ↗Explore all models