Smart Turn v3.2 CPU archive

An unchanged archival copy of Pipecat/Daily's smart-turn-v3.2-cpu.onnx, the CPU/int8 ONNX model from pipecat-ai/smart-turn-v3 at commit f766f81d3cfdf7737ac64aad813d91bbfd56bf93. The model was neither retrained nor converted for this mirror.

  • File size: 8,679,182 bytes
  • SHA-256: 2bb026316b14a660486a75b1733cd3fbab8c2fd0314dc9af7be49f8cca967e4f
  • Original file

Smart Turn was developed by Pipecat/Daily using OpenAI's Whisper Tiny encoder. The original model card, including dataset acknowledgements, is preserved byte-for-byte in UPSTREAM_README.md. No training or evaluation data is mirrored. No endorsement by Pipecat, Daily, OpenAI, or the upstream data contributors is implied.

Upstream declares Smart Turn BSD-2-Clause; see LICENSE. Whisper-related MIT and Apache-2.0 source declarations and retained license texts are documented in THIRD_PARTY_NOTICES.md.

See PROVENANCE.json and SHA256SUMS.json for source and integrity records.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support