tonhowtf/omniget · error

whisper-cli falhou

Error message

whisper-cli falhou: {}

What it means

whisper-cli exited with a non-zero status after transcription was attempted; the {} is the last non-progress stderr line captured while streaming 'progress =' lines to the progress reporter. Common causes are an unsupported/corrupt model file, missing audio input, or invalid CLI arguments.

Solutions

1. Read the captured stderr tail to identify the exact whisper-cli failure
2. Confirm the whisper model file exists, is complete, and matches the model flag used
3. Re-encode the audio to 16kHz WAV, the format whisper-cli expects, and retry

When it happens

Trigger: Thrown at src-tauri/omniget-core/src/core/tools/whisper.rs:389 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tonhowtf/omniget@8600b91f42 (2026-09-12). Data as JSON: /api/errors/28b68c13446022f8. Report an issue: GitHub.

Appendix: source

Thrown at src-tauri/omniget-core/src/core/tools/whisper.rs:389

        if let Some(err) = stderr {
            let mut lines = BufReader::new(err).lines();
            while let Ok(Some(line)) = lines.next_line().await {
                if let Some(rest) = line.trim().strip_prefix("progress =") {
                    if let Ok(pct) = rest.trim().trim_end_matches('%').trim().parse::<u64>() {
                        report(&p2, &id2, "transcribe", pct.min(100), Some(100), None);
                    }
                } else if !line.trim().is_empty() {
                    tail = line;
                }
            }
        }
        tail
    });
    let status = child.wait().await?;
    let tail = stderr_task.await.unwrap_or_default();
    let _ = tokio::fs::remove_file(&wav).await;
    if !status.success() {
        return Err(anyhow!("whisper-cli falhou: {}", tail));
    }
    let json_path = out_prefix.with_extension("json");
    let text = tokio::fs::read_to_string(&json_path).await?;
    let _ = tokio::fs::remove_file(&json_path).await;
    let (language, cues) = parse_whisper_json(&text)?;

    let out_dir = if opts.output_dir.trim().is_empty() {
        input
            .parent()
            .map(|p| p.to_path_buf())
            .unwrap_or_else(|| PathBuf::from("."))
    } else {
        PathBuf::from(opts.output_dir.trim())
    };
    std::fs::create_dir_all(&out_dir)?;
    let stem = input
        .file_stem()
        .map(|s| s.to_string_lossy().to_string())

View on GitHub (pinned to 8600b91f42)