Developers

Parexa REST API for text to speech and speech to text.

Authentication

Every request needs a key created on the API Keys page. Send it as a bearer token:

Authorization: Bearer px_live_...

Base URL: https://parexa.xyz (local: http://localhost:3000)

POST/api/v1/audio/speechtext → mp3

Synthesize speech. Returns raw audio (default audio/mpeg).

Field	Type	Description
text	string · required	Up to 4,000 characters.
voice	string	Edge neural voice, e.g. `en-IN-PrabhatNeural`.
rate	number 0.5–2	Speaking speed. Default 1.
pitch	number 0.5–2	Voice pitch. Default 1.
format	string	`mp3-48k` (default), `mp3-96k`, `opus-24k`.

curl https://parexa.xyz/api/v1/audio/speech \
  -H "Authorization: Bearer $PAREXA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text":"Hello from Parexa","voice":"en-US-AriaNeural"}' \
  --output speech.mp3

const res = await fetch("https://parexa.xyz/api/v1/audio/speech", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.PAREXA_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ text: "Hello from Parexa", voice: "en-US-AriaNeural" }),
});

const audio = Buffer.from(await res.arrayBuffer());

POST/api/v1/audio/transcriptionsaudio → text

Transcribe an audio file. Send multipart/form-data with the recording in the file field (max 25 MB).

Field	Type	Description
file	file · required	mp3, wav, m4a, ogg, webm. Max 25 MB.
language	string	Optional ISO-639-1 hint, e.g. `hi`.
response_format	json \| text	Default json.

curl https://parexa.xyz/api/v1/audio/transcriptions \
  -H "Authorization: Bearer $PAREXA_API_KEY" \
  -F "[email protected]" \
  -F "language=en"

{
  "text": "Hello from Parexa",
  "provider": "cloudflare",
  "language": "en"
}

GET/api/v1/voicesvoice catalog

List available voices. Filter with ?locale=en-IN or ?search=aria.

curl "https://parexa.xyz/api/v1/voices?locale=en-IN" \
  -H "Authorization: Bearer $PAREXA_API_KEY"

Limits & errors

· 60 requests per minute per key
· TTS text up to 4,000 characters per request
· Audio uploads up to 25 MB

Status	Meaning
401	Missing or invalid API key.
400 / 413 / 415	Bad input, too large, unsupported audio type.
422	No speech detected in the audio.
429	Rate limit exceeded.
502 / 503	Upstream engine failed or not configured.

Errors return { "error": { "message", "type" } }.