vercel/ai · error · InvalidArgumentError

The Gemini Live transcription API only supports 16kHz 16-bit

Error message

The Gemini Live transcription API only supports 16kHz 16-bit PCM input audio.

What it means

Error "The Gemini Live transcription API only supports 16kHz 16-bit PCM input audio." thrown in vercel/ai.

Source

Thrown at packages/google/src/transcription/google-transcription-model.ts:677

}

/** Parses a Google duration offset such as `"1s"` or `"9.400s"` to seconds. */
function parseOffsetSeconds(
  offset: string | undefined | null,
): number | undefined {
  if (offset == null) return undefined;
  const parsed = Number.parseFloat(offset);
  return Number.isFinite(parsed) ? parsed : undefined;
}

function validateLiveInputAudioFormat(
  inputAudioFormat: TranscriptionModelV4StreamOptions['inputAudioFormat'],
) {
  if (
    inputAudioFormat.type !== 'audio/pcm' ||
    (inputAudioFormat.rate != null && inputAudioFormat.rate !== 16000)
  ) {
    throw new InvalidArgumentError({
      argument: 'inputAudioFormat',
      message:
        'The Gemini Live transcription API only supports 16kHz 16-bit PCM input audio.',
    });
  }
}

const googleInteractionsWordAnnotationSchema = z.object({
  type: z.string().nullish(),
  text: z.string().nullish(),
  speaker: z.string().nullish(),
  start_offset: z.string().nullish(),
  end_offset: z.string().nullish(),
});

const googleInteractionsTranscriptionResponseSchema = z.object({
  status: z.string().nullish(),
  steps: z

View on GitHub (pinned to 69428b1f8b)

When it happens

Trigger: Thrown at packages/google/src/transcription/google-transcription-model.ts:677 when the library encounters an invalid state.

Common situations: See trigger scenarios.


AI-assisted analysis of vercel/ai@69428b1f8b (2026-08-30). Data as JSON: /api/errors/b43fc5e65ad22aeb. Report an issue: GitHub.