Speech to text API quickstart

One POST, base64 audio in, text out. Copy-paste examples.

One endpoint

POST https://speakvora.com/api/v1/listen with header x-api-key and a JSON body. language is optional; leave it out to detect the language.

curl

curl -X POST https://speakvora.com/api/v1/listen \
  -H "x-api-key: YOUR_KEY" -H "content-type: application/json" \
  -d "{\"audio_base64\": \"$(base64 < clip.wav | tr -d '\n')\", \"format\": \"wav\", \"language\": \"en\"}"

Python

import base64, requests

audio = base64.b64encode(open("clip.wav", "rb").read()).decode()
r = requests.post("https://speakvora.com/api/v1/listen",
                  headers={"x-api-key": "YOUR_KEY"},
                  json={"audio_base64": audio, "format": "wav", "language": "en"})
print(r.json()["text"])

Node

import fs from "node:fs";

const r = await fetch("https://speakvora.com/api/v1/listen", {
  method: "POST",
  headers: { "x-api-key": "YOUR_KEY", "content-type": "application/json" },
  body: JSON.stringify({ audio_base64: fs.readFileSync("clip.wav").toString("base64"), format: "wav", language: "en" }),
});
console.log((await r.json()).text);

Response

{ "text": "Your refund will arrive in three to five business days.", "seconds": 3, "language": "en" }

Silence or music returns empty text with a nonspeech field.

Errors

StatusCodeMeaning
400invalid_requestBad base64, unknown format, or audio under 1 KB or over 4.5 MB
401unauthorizedMissing or wrong key
402free_allowance_usedThe 60 free minutes are used: add a card
429rate_limitedToo many requests just now; wait the number of seconds in the answer
502upstream_unavailableSpeech recognition did not answer: retry

Longer recordings and live speech

· Stream live transcription over WebSocket

Cheapest speech to text API · Full reference

Questions

How do I transcribe an audio file with the API?

POST /api/v1/listen with the header x-api-key and a JSON body: audio_base64 (the file, base64 encoded), format (wav, mp3, m4a, ogg, flac or webm) and optionally language. The answer contains text, seconds and language.

What errors can I get?

400 invalid_request (bad base64, unknown format, or audio under 1 KB or over 4.5 MB), 401 unauthorized, 402 free_allowance_used or spend_cap_reached, 429 rate_limited, and 502 upstream_unavailable. Retry a 502 after a moment.

What if my file is larger than 4.5 MB?

Cut it into pieces (for example one to three minutes each, at a pause) and send them in order, then join the text. The guide on long recordings shows how.

Is there an SDK?

Not for speech to text yet: it is a single HTTP call, shown above in curl, Python and Node.

Does it cost anything to try?

No. The first 60 minutes are free with no credit card; then $0.0015 per minute.

What are the limits?

Up to 4.5 MB of audio per request (send longer recordings in pieces), formats wav, mp3, m4a, ogg, flac and webm. No word timestamps and no speaker labels yet. We have not published an accuracy benchmark, so test on a sample of your own recordings. Audio is sent through our server to a hosted speech-recognition service to be transcribed.