christopherthompson81/vibevoice-asr-streaming-7b-onnx
098
VibeVoice-ASR-Streaming 7B — ONNX export for Vernacula
An ONNX export of `microsoft/VibeVoice-ASR-Streaming-7B`, Microsoft's streaming speaker-attributed ASR model (Qwen2.5-7B decoder over a pair of causal audio tokenizers), for use as an ASR backend in Vernacula.
The model transcribes who said what as audio arrives, with no separate diarizer: it emits one text chunk per 2.93 s of audio and marks speaker turns inline.
- Conversion scripts: `scripts/vibevoice_streaming_export/`
- Investigation log: `docs/dev/vibevoice_asr_streaming_investigation.md`
Files
How it differs from the source checkpoint
- Deterministic audio encoding. Upstream samples the acoustic latents with Gaussian noise; this export uses the latent mean, which is bit-repeatable and sits inside the seed-to-seed variation of the original.
- INT8 decoder weights. Measured indistinguishable from float16 on transcript accuracy, while halving the weights.
- Shared, pre-allocated KV cache. VRAM is flat in recording length rather than growing, and throughput does not decay over a long file.
Requirements and limits
- ONNX Runtime with the CUDA execution provider.
GroupQueryAttentionand the float16 graphs are not supported on the CPU provider here. - Peak VRAM about 15.7 GB; real-time factor about 0.142 on an RTX 3090.
- The KV cache ceiling in
export-report.jsonbounds recording length and is set to the checkpoint's trained context (131,072 positions, roughly 2 hours of audio). The runtime refuses a longer recording rather than failing partway. Attention cost grows with the filled cache, so decoding slows as a long recording proceeds; VRAM does not. - Speaker attribution at this size is reliable — it separated two speakers correctly across a 30-minute file.
Licence
MIT, matching the source checkpoint.
