Hmbown/CodeWhale · error · anyhow::Error

`speech` requires provider = "xiaomi-mimo" (current: {}). Ru

Error message

`speech` requires provider = "xiaomi-mimo" (current: {}). Run with `--provider xiaomi-mimo` or set it in config.

What it means

The `speech` subcommand is wired only to the Xiaomi MiMo TTS API. Before building any request it asserts config.api_provider() == ApiProvider::XiaomiMimo and bails with the current provider's name plus the two remediation paths: the `--provider xiaomi-mimo` flag or a config change. No network I/O happens before this check.

Source

Thrown at crates/tui/src/lib.rs:7355

        default_speech_output_name, describe_speech_voice, encode_voice_clone_sample_data_uri,
        infer_speech_model, normalize_speech_format,
    };

    let SpeechArgs {
        text,
        output,
        output_dir,
        model,
        voice,
        instruction,
        voice_prompt,
        clone_voice,
        format,
        json: json_output,
    } = args;

    if config.api_provider() != ApiProvider::XiaomiMimo {
        bail!(
            "`speech` requires provider = \"xiaomi-mimo\" (current: {}). Run with `--provider xiaomi-mimo` or set it in config.",
            config.api_provider().as_str()
        );
    }

    if text.trim().is_empty() {
        bail!("Speech text cannot be empty");
    }
    let voice_is_data_uri = voice
        .as_deref()
        .map(str::trim)
        .is_some_and(|value| value.starts_with("data:audio/"));
    if clone_voice.is_some() && voice.is_some() {
        bail!("Use either --clone-voice or --voice for cloned voice data, not both");
    }
    let model = infer_speech_model(
        model.as_deref(),
        clone_voice.is_some() || voice_is_data_uri,

View on GitHub (pinned to 0c42157ee5)

Solutions

1. Add `--provider xiaomi-mimo` to the speech invocation
2. Or set provider = "xiaomi-mimo" in the config that speech runs under
3. Confirm the active provider afterwards with `codewhale doctor`

Example fix

// before
$ codewhale speech ...
Error: `speech` requires provider = "xiaomi-mimo" (current: deepseek). Run with `--provider xiaomi-mimo` or set it in config.

// after
$ codewhale speech --provider xiaomi-mimo ...

# pin the provider explicitly so config drift cannot break speech jobs
exec codewhale speech --provider xiaomi-mimo "$@"

When it happens

Trigger: Running `codewhale speech ...` while the active config's api_provider resolves to anything other than xiaomi-mimo (e.g. a chat-provider default). The check is the first thing after argument destructuring.

Common situations: A config.toml defaulting to the chat provider for coding work; switching providers globally and forgetting speech is MiMo-only; a fresh machine where the default provider was never changed.

Related errors

• invalid value '{provider}' for '--provider <PROVIDER>': expe
• speech synthesis requires provider 'xiaomi-mimo' (current: {
• Unrecognized --provider {provider_arg:?}. Known providers: {
• Use either --clone-voice or --voice for cloned voice data, n
• speech requires a TTS model (examples: {}); got {model}

AI-assisted analysis of Hmbown/CodeWhale@0c42157ee5 (2026-08-20). Data as JSON: /api/errors/c85cdbbdda5a7f06. Report an issue: GitHub.