Mintplex-Labs/anything-llm · warning · Error

Invalid audio file: ${error.message}

Error message

Invalid audio file: ${error.message}

What it means

Wrapper thrown when #validateAudioFile raises (errors 25/26/27). #convertToWavAudioData catches the validation error, logs it, and re-throws with an "Invalid audio file:" prefix. This is the surfaced form of the three validation errors.

Source

Thrown at collector/utils/WhisperProviders/localWhisper.js:88

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

      const outputFile = path.resolve(outFolder, `${v4()}.wav`);
      const success = await ffmpeg.convertAudioToWav(sourcePath, outputFile);
      if (!success)
        throw new Error(
          "[Conversion Failed]: Could not convert file to .wav format!"
        );

      buffer = fs.readFileSync(outputFile);
      fs.rmSync(outputFile);

      const wavFile = new wavefile.WaveFile(buffer);
      try {
        this.#validateAudioFile(wavFile);
      } catch (error) {
        this.#log(`Audio validation failed: ${error.message}`);
        throw new Error(`Invalid audio file: ${error.message}`);
      }

      // Although we use ffmpeg to convert to the correct format (16k hz 32f),
      // different versions of ffmpeg produce different results based on the
      // environment. To ensure consistency, we convert to the correct format again.
      wavFile.toBitDepth("32f");
      wavFile.toSampleRate(16000);

      let audioData = wavFile.getSamples();
      if (Array.isArray(audioData)) {
        if (audioData.length > 1) {
          const SCALING_FACTOR = Math.sqrt(2);

          // Merge channels into first channel to save memory
          for (let i = 0; i < audioData[0].length; ++i) {
            audioData[0][i] =
              (SCALING_FACTOR * (audioData[0][i] + audioData[1][i])) / 2;
          }

View on GitHub (pinned to 526360e320)

Solutions

1. Parse the inner message to determine which validation failed and act (re-encode, split, reject).
2. Re-encode the source through ffmpeg to 16 kHz mono and retry.
3. Split long files into <4 h chunks.

Example fix

// before
throw new Error(`Invalid audio file: ${error.message}`);

// after — typed error for programmatic handling
const err = new Error(`Invalid audio file: ${error.message}`);
err.cause = error;
err.code = error.message.includes("sample rate") ? "LOW_SAMPLE_RATE"
  : error.message.includes("duration") ? "TOO_LONG" : "TOO_MANY_SAMPLES";
throw err;

try { await localWhisper.processFile(filePath, name); }
catch (e) {
  if (e.message.startsWith("Invalid audio file:")) {
    const reason = e.message.slice("Invalid audio file:".length).trim();
    /* branch on reason: re-encode, split, or reject */
  }
  throw e;
}

When it happens

Trigger: Any of the three validation failures (low sample rate, >4 h duration, projected samples too high) inside #convertToWavAudioData triggers the catch, which re-wraps the message.

Common situations: Bad/corrupt audio; a pre-existing WAV not normalized by ffmpeg; extremely long recordings.

Related errors

• No base URL was set.
• Audio file sample rate is too low for accurate transcription
• Audio file duration exceeds maximum limit of 4 hours.
• Audio file exceeds maximum allowed length.
• No OpenAI API key was set.

AI-assisted analysis of Mintplex-Labs/anything-llm@526360e320 (2026-08-13). Data as JSON: /api/errors/5aef7473d6d8325b. Report an issue: GitHub.