Anvia Docs

Import from @anvia/core or @anvia/core/transcription.

TranscriptionModel

interface TranscriptionModel<RawResponse = unknown> {
  readonly provider?: string;
  readonly defaultModel?: string;
  transcription(request: TranscriptionRequest): Promise<TranscriptionResponse<RawResponse>>;
}

Purpose: provider-neutral audio transcription contract.

Return behavior: providers return normalized text and the provider response as rawResponse.

Request Builder

type TranscriptionRequest = {
  data: Uint8Array;
  filename: string;
  language?: string;
  prompt?: string;
  temperature?: number;
  additionalParams?: JsonValue;
};

type TranscriptionResponse<RawResponse = unknown> = {
  text: string;
  rawResponse: RawResponse;
};

class TranscriptionRequestBuilder<Model extends TranscriptionModel = TranscriptionModel> {
  data(data: Uint8Array | ArrayBuffer): this;
  filename(filename: string): this;
  language(language: string): this;
  prompt(prompt: string): this;
  temperature(temperature: number): this;
  additionalParams(additionalParams: JsonValue): this;
  build(): TranscriptionRequest;
  send(): Promise<TranscriptionResponse>;
}

const response = await transcriptionRequest(model)
  .data(audioBytes)
  .filename("meeting.mp3")
  .language("en")
  .prompt("Transcribe exactly.")
  .temperature(0)
  .send();

Purpose: Chainable builder for transcription requests.

Defaults: filename: "file" and empty data.

Notable errors: .build() and .send() throw when data is empty. V1 does not include a filesystem loadFile() helper; callers pass bytes.