Transcribe Audio

Summary: Local speech‑to‑text that supports streaming or batch, producing normalized transcripts for downstream LLMs or UIs.

Use This When

• You need speech‑to‑text for voice UIs, captioning, or command routing.
• You’re handing transcriptions to LLMs for reasoning or summarization.

What It Does

• Converts audio frames to text with a local model; supports streaming or batch.
• Handles basic chunking and normalization heuristics.

Works Best With

• VAD (to gate silence), text analytics, and TTS for round‑trip agents.

Caveats

• Noise and domain mismatch reduce accuracy; microphone and environment matter.