`experimental_streamTranscribe()`

experimental_streamTranscribe is an experimental feature.

Streams a transcript from live raw audio using a transcription model with streaming support.

import { experimental_streamTranscribe as streamTranscribe } from 'ai';
import { openai } from '@ai-sdk/openai';

const result = streamTranscribe({
  model: openai.transcription('gpt-realtime-whisper'),
  audio: audioStream, // ReadableStream<Uint8Array | string>
  inputAudioFormat: { type: 'audio/pcm', rate: 24000 },
});

for await (const part of result.fullStream) {
  if (part.type === 'transcript-delta') {
    process.stdout.write(part.delta);
  }
}

console.log(await result.text);

Import

import { experimental_streamTranscribe as streamTranscribe } from "ai"

API Signature

model:

TranscriptionModelV4

audio:

ReadableStream<Uint8Array | string>

inputAudioFormat:

{ type: string; rate?: number }

providerOptions?:

Record<string, JSONObject>

abortSignal?:

AbortSignal

headers?:

Record<string, string>

includeRawChunks?:

boolean

fullStream:

AsyncIterableStream<TranscriptionStreamPart>

text:

Promise<string>

segments:

Promise<Array<{ text: string; startSecond: number; endSecond: number }>>

language:

Promise<string | undefined>

durationInSeconds:

Promise<number | undefined>

warnings:

Promise<Warning[]>

responses:

Promise<Array<TranscriptionModelResponseMetadata>>

providerMetadata:

Promise<Record<string, JSONObject>>

The result promises settle as the stream is consumed. If you stop consuming fullStream early (e.g. break out of the loop), the underlying provider connection is closed and pending result promises reject.