Mintplex-Labs/anything-llm · error · Error

Failed to transcribe audio.

Error message

Failed to transcribe audio.

What it means

Thrown by the Telegram voice-transcription path when collector.parseDocument(filename) returns no usable result (result.success falsy or no documents). The audio buffer was already written to the collector hotdir and parsed; failure here means transcription itself failed or returned empty. The thrown message uses result.reason if provided, else the generic string.

Source

Thrown at server/utils/telegramBot/utils/media.js:61

 * @param {string} [mimeType] - The MIME type of the audio (e.g., "audio/ogg")
 * @returns {Promise<string>}
 */
async function transcribeAudio(audioBuffer, mimeType = "audio/ogg") {
  const fs = require("fs");
  const path = require("path");
  const { CollectorApi } = require("../../collectorApi");
  const { hotdirPath } = require("../../files");

  if (!fs.existsSync(hotdirPath)) fs.mkdirSync(hotdirPath, { recursive: true });

  const ext = getExtensionFromMime(mimeType);
  const filename = `telegram-voice-${Date.now()}${ext}`;
  fs.writeFileSync(path.join(hotdirPath, filename), audioBuffer);

  const collector = new CollectorApi();
  const result = await collector.parseDocument(filename);
  if (!result?.success || !result.documents?.length) {
    throw new Error(result?.reason || "Failed to transcribe audio.");
  }
  return result.documents[0].pageContent;
}

/**
 * Parse a document buffer and extract its text content.
 * Writes the document to the collector hotdir and runs it through
 * the collector's parse pipeline.
 * @param {Buffer} documentBuffer
 * @param {string} originalFilename - The original filename with extension
 * @returns {Promise<{text: string, filename: string}>}
 */
async function documentToText(documentBuffer, originalFilename) {
  const fs = require("fs");
  const path = require("path");
  const { CollectorApi } = require("../../collectorApi");
  const { hotdirPath } = require("../../files");

View on GitHub (pinned to 526360e320)

Solutions

1. Check the collector logs for the underlying transcription error (often a model/API failure).
2. Verify the transcription backend config (API key, model name, endpoint) in collector settings.
3. Ensure ffmpeg/decoding deps are present in the collector image for the incoming codec.
4. Reply to the user asking them to resend as a common format (.mp3, .ogg) if the codec is exotic.

Example fix

// before
const text = (await collector.parseDocument(filename)).documents[0].pageContent;

// after
const result = await collector.parseDocument(filename);
if (!result?.success || !result.documents?.length)
  throw new Error(result?.reason || 'Failed to transcribe audio.');
const text = result.documents[0].pageContent;

const result = await collector.parseDocument(filename);
if (!result?.success || !result.documents?.length)
  throw new Error(result?.reason || 'Failed to transcribe audio.');

try {
  const text = await transcribeTelegramAudio(bot, fileId, mimeType);
} catch (e) {
  if (e.message === 'Failed to transcribe audio.')
    return ctx.reply('Transcription failed; please resend the voice message.');
  throw e;
}

When it happens

Trigger: A voice message in an unsupported audio codec; the collector's transcription backend (e.g. whisper) failing or misconfigured; an empty/corrupt audio buffer; the collector returning success=false with a reason.

Common situations: Whisper API key missing or invalid; collector container out of memory during transcription; audio file truncated during Telegram transfer; unsupported .oga variant the transcoder cannot decode.

Related errors

• Audio conversion failed.
• Document processing is unavailable. The collector service is
• Failed to parse document: ${originalFilename}
• Failed to sync YouTube video transcript. ${reason}
• No base URL was set.

AI-assisted analysis of Mintplex-Labs/anything-llm@526360e320 (2026-08-13). Data as JSON: /api/errors/75db60b431d9e7fb. Report an issue: GitHub.