What causes "MiMo TTS returned no audio" in decolua/9router?

Log the full response JSON to inspect choices and any embedded error message

decolua/9router · error · Error

MiMo TTS returned no audio

Error message

MiMo TTS returned no audio

What it means

Thrown when the MiMo upstream returns HTTP OK but the response JSON has no choices[0].message.audio.data — i.e., the API accepted the request but returned no audio payload. The handler prefers data.error.message if present to hint at the cause.

Source

Thrown at open-sse/handlers/ttsProviders/xiaomi-mimo.js:59

      audio: {
        format: "wav",
        voice: voiceId || DEFAULT_VOICE,
      },
    }),
  });

  const rawText = await res.text();
  let data = {};
  if (rawText) {
    try { data = JSON.parse(rawText); } catch { data = {}; }
  }

  if (!res.ok) {
    throw new Error(data?.error?.message || rawText || `MiMo TTS error (${res.status})`);
  }

  const audio = data?.choices?.[0]?.message?.audio?.data;
  if (!audio) throw new Error(data?.error?.message || "MiMo TTS returned no audio");

  return {
    base64: audio,
    format: data?.choices?.[0]?.message?.audio?.format || "wav",
  };
}

View on GitHub (pinned to 90b52e06ff)

Solutions

Log the full response JSON to inspect choices and any embedded error message
Verify the account has MiMo TTS entitlement enabled
Shorten/sanitize input text to rule out moderation or length limits
Check for MiMo API version changes to the audio response shape
Retry once to rule out a transient upstream hiccup

Defensive patterns

Strategy: fallback

Validate before calling

if (typeof text !== "string" || text.length === 0 || text.length > MAX_MIMO_INPUT) throw new Error("MiMo TTS: invalid input text");

Type guard

function mimoResponseHasAudio(d) {
  const a = d?.choices?.[0]?.message?.audio;
  return typeof a?.data === "string" && a.data.length > 0;
}

Try / catch

try {
  return await synthesizeMiMo(text, model, key, style, language);
} catch (e) {
  if (e.message === "MiMo TTS returned no audio") {
    logger.warn("mimo tts: empty audio in 200 response", { model });
    return fallbackTts(text);
  }
  throw e;
}

Prevention

Sanitize/shorten input text to avoid moderation and length rejections
Log the full 200-response body at debug level to catch shape changes
Verify TTS entitlement is active for the Xiaomi account
Fall back to a secondary TTS provider when audio is empty

When it happens

Trigger: MiMo returns 200 with an empty choices array, a text-only message, or an audio object missing data (e.g. content-filtered input, silent generation failure, or API response shape change).

Common situations: Input text blocked by content moderation, account without TTS entitlement returning an empty success, MiMo API version updating the audio field location, or extremely long input being truncated to no audio.

Related errors

AI-assisted analysis of decolua/9router@90b52e06ff (2026-08-30). Data as JSON: /api/errors/62340f9896dbaa2e. Report an issue: GitHub.