Speech 02 HD is a MiniMax model exposed on the Huzz API with a single Huzz key and one consistent, OpenAI-compatible interface. Capabilities: Audio generation, Voice cloning. It supports streaming responses. Long-running jobs run asynchronously with polling. Call it at https://api.huzz.ai using Authorization: Bearer $HUZZ_API_KEY — no per-provider accounts or SDKs required.
FamilyCapabilitiesStreamingAsyncPricing
MiniMaxAudio generation, Voice cloningYesYesUsage-based

​
Endpoints

MethodPathCapability
POST/api/v3/minimax/speech-02-hd/text-to-audioAudio generation
POST/api/v3/minimax/speech-02-hd/voice-cloneVoice cloning

​
Code samples

Authenticate with Authorization: Bearer $HUZZ_API_KEY — get a key at huzz.ai/register. 
# 1) Submit the job
curl https://api.huzz.ai/api/v3/minimax/speech-02-hd/text-to-audio \
  -H "Authorization: Bearer $HUZZ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A timelapse of city lights at night"}'

# 2) Poll for the result
curl https://api.huzz.ai/api/v3/predictions/PREDICTION_ID/result \
  -H "Authorization: Bearer $HUZZ_API_KEY"
Long-running jobs return a prediction id. See Async predictions for the full submit-and-poll lifecycle.