| --- |
| language: |
| - en |
| license: openmdw-1.1 |
| license_link: LICENSE.md |
| pipeline_tag: audio-classification |
| base_model: |
| - nvidia/nemotron-3.5-asr-streaming-0.6b |
| - FluidInference/parakeet-redux-coreml |
| tags: |
| - decision-model |
| - turn-taking |
| - endpoint-detection |
| - barge-in |
| - backchannel |
| - full-duplex |
| - streaming |
| - on-device |
| - coreml |
| - core-ai |
| - mlx |
| - gru |
| - audio |
| --- |
| |
| # tyger-models |
|
|
| **On-device delivery tiers for [TYGER](https://huggingface.co/nineninesix/tyger) |
| β an open-weight speech decision model for full-duplex voice agents β plus |
| `TYGER-Step-Student`, a distilled turn-taker that reads |
| [Parakeet Redux](https://huggingface.co/FluidInference/parakeet-redux-coreml) |
| features instead of Nemotron's.** |
|
|
| TYGER knows when to respond, keep listening, yield, and backchannel. This repo |
| is the shipping form: the same decisions, at device scale, in the formats |
| iOS and macOS load directly. It holds weights and a download manifest, no |
| inference code. |
|
|
| ## What's here |
|
|
| | Path | Model | Format | Size | |
| | --- | --- | --- | ---: | |
| | `coreai/` | TYGER chain (encoder + turn heads) | Core AI `.aimodel`, fp16 | ~0.77 GB | |
| | `coreml/` | TYGER chain | Core ML `.mlpackage` (encoder fp16, step fp32) | ~0.86 GB | |
| | `mlx/` | TYGER chain | MLX int8-g16 `.safetensors` (pre-27 macOS) | 1.58 GBΒΉ | |
| | `pathc/` | **TYGER-Step-Student** (turn heads only) | Core ML `.mlpackage`, fp32 | **56 MB** | |
| | `manifest.json` | β | delivery manifest (sha256 per file) | β | |
|
|
| ΒΉ the artifact stores dequantized fp32 weights. |
|
|
| The tiers are interchangeable exports of |
| [nineninesix/tyger](https://huggingface.co/nineninesix/tyger)'s |
| `model.safetensors` (638M Nemotron + 14M GRU heads); see that model page for |
| the architecture, event semantics (endpoint / barge-in / backchannel), guards |
| and thresholds. `pathc/` is a different animal and is documented below. |
|
|
| ## TYGER-Step-Student (`pathc/`) |
|
|
| The turn-taker alone β **14M parameters** (a causal GRU, 3 layers, width 768, |
| plus the input projections and five output heads; the same architecture as |
| TYGER's GRU) β retrained to read **Parakeet Redux** encoder features instead of |
| Nemotron layer-15 features. |
|
|
| **Why:** a voice agent that runs Parakeet for transcription already computes |
| this model's input. The student shares that encoder pass, so the turn model's |
| delivery cost collapses from a second ~0.8 GB encoder chain to **56 MB of |
| heads**, identical across device tiers. Total first-use voice download with |
| ASR + TTS + this model: ~375 MB. |
|
|
| **Input / output** (per 320 ms hop, same contract as TYGER's step): |
|
|
| | | | |
| | --- | --- | |
| | Input features | Parakeet Redux `Preprocessor` + `Encoder` output: 188 rows @ 12.5 Hz Γ 1,024 dims per stream (user + agent) | |
| | Tap contract | 15 s right-aligned, zero-padded window (mel_length 1501); run both encoder models **CPU-only** (ANE/GPU drifts from training features and is not faster) | |
| | Per-frame extras | frame energy + the previous state | |
| | State | GRU carried state `(3, 1, 768)`, ~9 KiB per conversation | |
| | Output, per frame | the turn state logits (listen, hold, speak, backchannel) + hazard, yield, backchannel and endpoint scores β 8 numbers | |
| |
| ### Training |
| |
| Distilled from **TYGER** (the fp32 Nemotron chain as teacher) on dual-channel |
| English conversation β the |
| [dualturn-otospeech-turn-taking](https://huggingface.co/datasets/anyreach-ai/dualturn-otospeech-turn-taking) |
| corpora, codec-round-tripped as in calls: 96 sessions (~4.8 h of frames) for |
| training, 24 sessions (~1.3 h) held out for evaluation. Targets are the |
| teacher's per-frame 8-vectors; boundary frames are weighted 3Γ (MSE, equal |
| group weighting). 250 + 200 epochs, AdamW with cosine decay. One training |
| seed. The rules that turn scores into events (warm-up frames, speech gate, |
| silence fallback) live app-side and are unchanged from tyger-s2s's. |
| |
| ### Evaluation (in use) |
| |
| Scored by replaying held-out conversations through the full chain and |
| comparing *behavior* β did the model act at real turn-ends β not frame identity |
| with the teacher. 24 validation sessions, 156 human turn-ends: |
| |
| | Metric | Teacher (TYGER) | Student | |
| | --- | ---: | ---: | |
| | Real turn-ends caught | 87% | **91%** (142/156) | |
| | False endpoints / min | 6.2 | 6.7 | |
| | Backchannel precision / recall | 20% / 48% | 22% / 44% | |
| |
| Core ML export parity against the training-time framework: max |Ξ| 1e-5 over |
| 400 carried-state frames. |
| |
| **Limits:** English only. The evaluation is recorded conversation replayed |
| through the pipeline β not a live voice-agent benchmark β and one training |
| seed. Frame-identity with the teacher is ~48%: the student learns the |
| teacher's behavior, not its every twitch. Backchannel recall is modest for |
| teacher and student alike. False-endpoint rates are corpus- and |
| threshold-dependent; tune thresholds in the app. |
| |
| ## Intended use |
| |
| Turn-taking control for full-duplex voice agents on device: deciding when the |
| user has finished, when they cut in, when a backchannel fits, and when to |
| hold the floor. The agent above it sees only text and turn events. |
| |
| Not for transcription (use the ASR alongside it), not for safety-critical |
| decisions, not for non-English speech. |
| |
| ## Delivery format |
| |
| `manifest.json` pins every file (`name`, `asset`, `sha256`, `bytes`) with a |
| `url_base`. Files are flattened (`/` β `__`); `name` keeps the bundle-relative |
| path so a client can recreate the `.mlpackage`/`.aimodel` directory trees and |
| verify each byte before installing. The format is what an app-side |
| download-on-first-use store consumes: fetch the manifest, fetch only the |
| active tier's files. |
| |
| ## License |
| |
| The `coreai/`, `coreml/` and `mlx/` tiers are exports of |
| [TYGER](https://huggingface.co/nineninesix/tyger), whose weights and files are |
| **OpenMDW-1.1**; NVIDIA's original Nemotron card, figures and license terms |
| travel with that repo (`upstream/`, `licenses/`). This repo's tiers and the |
| `pathc/` student are distributed under the same **OpenMDW-1.1** terms. See |
| [LICENSE.md](LICENSE.md) and [NOTICE.md](NOTICE.md). |
| |
| `TYGER-Step-Student` was trained by distillation from TYGER on the |
| dualturn-otospeech-turn-taking corpora; see that dataset's own terms for the |
| training data. Its feature extractor is |
| [Parakeet Redux](https://huggingface.co/FluidInference/parakeet-redux-coreml) |
| (used, not redistributed here). |
| |
| ## Provenance & acknowledgments |
| |
| - **TYGER** β [nineninesix/tyger](https://huggingface.co/nineninesix/tyger) |
| (weights), [nineninesix-ai/tyger-s2s](https://github.com/nineninesix-ai/tyger-s2s) |
| (Apache-2.0 runtime the rules port follows). Thanks to NineNineSix. |
| - **Nemotron 3.5 ASR Streaming 0.6B** β |
| [nvidia/nemotron-3.5-asr-streaming-0.6b](https://huggingface.co/nvidia/nemotron-3.5-asr-streaming-0.6b), |
| unchanged upstream. |
| - **Parakeet Redux** β |
| [FluidInference/parakeet-redux-coreml](https://huggingface.co/FluidInference/parakeet-redux-coreml). |
| - **Training data** β The Agentic Data Company's |
| [Open Yap 1K](https://theagenticdatacompany.com/open-yap-1k) (TYGER's |
| training corpus) and the |
| [dualturn-otospeech-turn-taking](https://huggingface.co/datasets/anyreach-ai/dualturn-otospeech-turn-taking) |
| corpora (the student's). |
| |
| ## Citation |
| |
| If you use these tiers, please cite TYGER and its training data: |
| |
| ```bibtex |
| @misc{tyger_2026, |
| author = {{NineNineSix}}, |
| title = {{TYGER}: Speech Recognition and Duplex Turn Control}, |
| year = {2026}, |
| publisher = {NineNineSix}, |
| url = {https://huggingface.co/nineninesix/tyger} |
| } |
| |
| @misc{openyap_2026, |
| author = {{The Agentic Data Company}}, |
| title = {{Open Yap 1K}: 1,000 hours of full-duplex natural conversation}, |
| year = {2026}, |
| url = {https://theagenticdatacompany.com/open-yap-1k} |
| } |
| ``` |
| |
