chenhg5/cc-connect · error

gemini stt API %d: %s

Error message

gemini stt API %d: %s

What it means

Thrown by GeminiSTT.Transcribe in core/speech.go when the Gemini API returns a non-200 HTTP status. The error embeds the status code and the raw response body (Gemini error JSON typically explains invalid API key, bad model, quota exhaustion, or malformed request).

Source

Thrown at core/speech.go:277

	if err != nil {
		return "", fmt.Errorf("gemini stt: create request: %w", err)
	}
	req.Header.Set("Content-Type", "application/json")
	req.Header.Set("x-goog-api-key", g.APIKey)

	resp, err := g.Client.Do(req)
	if err != nil {
		return "", fmt.Errorf("gemini stt: request: %w", err)
	}
	defer resp.Body.Close()

	body, err := io.ReadAll(resp.Body)
	if err != nil {
		return "", fmt.Errorf("gemini stt: read response: %w", err)
	}

	if resp.StatusCode != http.StatusOK {
		return "", fmt.Errorf("gemini stt API %d: %s", resp.StatusCode, string(body))
	}

	var result struct {
		Candidates []struct {
			Content struct {
				Parts []struct {
					Text string `json:"text"`
				} `json:"parts"`
			} `json:"content"`
		} `json:"candidates"`
	}
	if err := json.Unmarshal(body, &result); err != nil {
		return "", fmt.Errorf("gemini stt: parse response: %w", err)
	}
	if len(result.Candidates) == 0 || len(result.Candidates[0].Content.Parts) == 0 {
		return "", fmt.Errorf("gemini stt: empty response")
	}

View on GitHub (pinned to 4000b2338a)

Solutions

1. Read the status code and body in the error: 401/403 -> fix the API key; 404 -> fix the model name; 400 -> check audio format/MIME; 429 -> wait or raise quota.
2. Verify the Gemini API key is set and valid in config.toml/env.
3. Confirm the configured model exists for your account/region (Google AI Studio model list).
4. Convert audio to a supported format (mp3) via ConvertAudioToMP3.
5. For 429/5xx, retry with exponential backoff.

Example fix

// before
model = "gemini-pro" // does not accept audio -> 400
// after
model = "gemini-1.5-flash" // audio-capable model, valid API key

if g.APIKey == "" { return errors.New("gemini: missing API key") }
if !strings.Contains(g.Model, "gemini") { return fmt.Errorf("gemini: suspicious model name %q", g.Model) }

text, err := stt.Transcribe(ctx, audio)
if err != nil {
    if strings.Contains(err.Error(), "gemini stt API 429") {
        return "", ErrRateLimited // back off and retry later
    }
    if strings.Contains(err.Error(), "gemini stt API 40") {
        return "", ErrConfigInvalid // key/model/audit request problem
    }
    return err
}

When it happens

Trigger: The POST to {BaseURL}/models/{model}:generateContent responds with any status other than 200: 400 invalid request/unsupported audio MIME, 401/403 invalid or missing API key, 404 unknown model name, 429 quota exceeded, 500/503 Gemini outage.

Common situations: Missing or revoked GEMINI/Google AI Studio API key; model name typo (e.g. gemini-1.5-flash vs deleted preview version); audio MIME type not supported by the model; free-tier quota exhausted; regional blocking of the API.

Understand the failure class

Background: "API error: {status}" and "HTTP 401/403/404/429/5xx" errors: non-2xx HTTP responses explained — this error's family across 27 libraries.

Related errors

• qwen asr API %d: %s
• gemini stt: request: %w
• gemini stt: create request: %w
• gemini stt: read response: %w
• range chunk retries exhausted

AI-assisted analysis of chenhg5/cc-connect@4000b2338a (2026-09-06). Data as JSON: /api/errors/8e7b1886c1fc6f73. Report an issue: GitHub.