Audio

Turn text into speech, and speech into text. Both endpoints are synchronous.

Text-to-speech

Generate spoken audio from text. The response body is the audio file itself.

POST https://api.heyaskr.ai/v1/audio/speech

string

required

Audio model id, e.g. elevenlabs-v3.

string

required

The text to speak.

string

Voice id, e.g. rachel (model dependent).

string

default:"mp3"

Output format, e.g. mp3, wav.

curl https://api.heyaskr.ai/v1/audio/speech \
  -H "Authorization: Bearer ASKR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "elevenlabs-v3",
    "voice": "rachel",
    "input": "Crypto in. Intelligence out."
  }' --output speech.mp3

The response is the raw audio file. The cost of the generation is returned in the X-Askr-Cost response header.

Speech-to-text

Transcribe an audio file to text. Send the file as multipart/form-data.

POST https://api.heyaskr.ai/v1/audio/transcriptions

string

required

Transcription model id, e.g. whisper-1.

file

required

The audio file to transcribe.

string

ISO code to hint the language (optional).

curl https://api.heyaskr.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer ASKR_API_KEY" \
  -F model="whisper-1" \
  -F file="@meeting.mp3"

Response

string

The transcribed text (transcriptions only).

number

Cost of the request in USD.

{ "text": "Welcome to the meeting...", "cost": 0.006 }

Model ids are examples, so confirm the exact id and current price against the live /v1/models list.

Overview

Inference

Accounts & billing

Text-to-speech

Speech-to-text

Response

​Text-to-speech

​Speech-to-text

​Response

Text-to-speech

Speech-to-text

Response