Speech to Text

Transcribes speech from either an uploaded audio file or a public audio URL. Submit multipart form data with exactly one audio source: content or url.

POST /v1/transcriptions

Speech to Text

cURL

curl --request POST \
  --url https://api.spitch.app/v1/transcriptions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form 'content=<string>' \
  --form 0.content='@example-file'

200

{
  "request_id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "text": "<string>",
  "segments": [
    {
      "text": "<string>",
      "start": 123,
      "end": 123
    }
  ],
  "timestamps": [
    {
      "text": "<string>",
      "start": 123,
      "end": 123
    }
  ],
  "detected_language": "<string>"
}

Authorizations

Authorization · string · header · required

Authenticate with Authorization: Bearer <token>. The service accepts JWTs, API keys, and guest tokens through this bearer token header.

Body

multipart/form-data

Multipart form data. Provide exactly one audio source: an uploaded file or a public URL.

Option 1

Option 2

content · file · required

Audio file to transcribe.

language · string

Optional ISO 639 language code for the spoken audio. Omit it to let the service use automatic language handling.

special_words · string

Optional comma-separated words to bias recognition toward domain-specific names or terms.

timestamp · enum<string> · default: none

Timestamp granularity for returned segments. Use none, sentence, or word.

Available options: sentence, word, none

Response

200

application/json

Successful Response

request_id · string<uuid> · required

text · string · required

segments · Segment · object[] | null

Show child attributes

timestamps · Segment · object[] | null

Show child attributes

detected_language · string | null