Thanabordee/Qwen3-ASR-0.6B-Thai-Engram
automatic-speech-recognition model by Thanabordee
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
A token-level bigram/trigram memory (Engram) injected into a frozen Qwen/Qwen3-ASR-0.6B, trained for Thai speech recognition. On a fixed 300-clip held-out set it reduced corpus character error rate from 17.8051 % to 9.0448 %. This repository holds a delta, not a full model. The base Qwen3-ASR-0.6B weights are not duplicated here — they are fetched from the Hub. The only trained parameters are the 8,249,344 Engram parameters in engramstep000750.pt (32 MiB). A hash-based N-gram lookup memory, not an adapter or a LoRA. It reads the token IDs of the decoder input, hashes every bigram and trigram through 16 independent heads into 10,000 / 2,000 buckets, and turns the retrieved vectors into a residual: It is injected after the FFN residual of decoder layer 2 only. During inference the residual is applied to the last prompt position and every generated position; the earlier prompt positions are masked out. keyproj, valueproj, and shortconv…
Read the full model card ↗ · Preview checked 2026-11-05T22:55:47.475Z
Inside the original model card — Document outline
- Qwen3-ASR-0.6B + Thai N-gram Engram
- What the Engram is
- Results
- Usage
- Training
- Limitations
- Licensing — read this
- Citation
- ภาษาไทย
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the model card
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: cc-by-sa-4.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
Thanabordee/Qwen3-ASR-0.6B-Thai-Engram is a automatic-speech-recognition model repository published by Thanabordee on Hugging Face. A library was not reported.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- automatic-speech-recognition
- Library
- Not reported
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- cc-by-sa-4.0
- Access
- Not gated by Hugging Face
- Created
- 2026-09-24T06:36:58.000Z
- Last modified
- 2026-09-24T06:37:10.000Z
Compatibility and lineage
Reported languages: th
Reported base models: Qwen/Qwen3-ASR-0.6B
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-09-24T06:37:32.927Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.
Related models
jonatasgrosman/wav2vec2-large-xlsr-53-japanese