Skip to main content
Transcribe an audio payload. The router picks the best STT provider for your (language, vertical, optimizeFor) and fails over automatically. 
const { text, provider, confidence } = await speko.transcribe(audio, {
  language: 'es-MX',
  vertical: 'healthcare',
});

​
Signature

speko.transcribe(
  audio: Uint8Array,
  options: TranscribeOptions,
  abortSignal?: AbortSignal,
): Promise<TranscribeResult>

​
Parameters

​
audio: Uint8Array

Raw audio bytes. Default MIME is audio/wav; set options.contentType if you send something else. Providers handle resampling and format conversion downstream — you don’t have to match a specific sample rate.

​
options: TranscribeOptions

Extends RoutingIntent:
FieldTypeDescription
languagestring (BCP-47)e.g. "en", "es-MX", "ja-JP".
vertical'general' | 'healthcare' | 'finance' | 'legal'Vertical label used by the router.
optimizeFor'balanced' | 'accuracy' | 'latency' | 'cost'?Bias the weighted score. Defaults to the server-side default (currently balanced).
contentTypestring?MIME type for the body. Defaults to audio/wav.
constraintsPipelineConstraints?Allow-list constraints (see Types).

​
abortSignal?: AbortSignal

Cancel an in-flight request. Composed with the client-level timeout.

​
Returns

​
TranscribeResult

FieldTypeDescription
textstringTranscribed text.
providerstringUpstream provider that ran the request (e.g. deepgram, openai).
modelstringProvider-specific model identifier.
confidencenumber | nullModel-reported confidence when available, else null.
failoverCountnumberHow many providers were tried before this one succeeded.
scoresRunIdstring | nullID of the scoring run that selected this provider — useful for joining to benchmark data.

​
Wire format

The SDK sends the audio as the raw HTTP body with:
  • Content-Type: from options.contentType (default audio/wav).
  • X-Speko-Intent: JSON-serialized { language, vertical, optimizeFor? }.
  • X-Speko-Constraints: JSON-serialized options.constraints (only if set).
The response is a JSON TranscribeResult.

​
Example: non-default MIME

import { readFile } from 'node:fs/promises';

const audio = await readFile('./call.ogg');
const result = await speko.transcribe(audio, {
  language: 'en',
  vertical: 'general',
  contentType: 'audio/ogg',
  optimizeFor: 'accuracy',
});

​
Example: restrict provider pool

const result = await speko.transcribe(audio, {
  language: 'en',
  vertical: 'healthcare',
  constraints: {
    allowedProviders: { stt: ['deepgram', 'assemblyai'] },
  },
});
The router still ranks candidates by benchmark score and only picks from the allow-list.