CoolFace
Modelpublic

mmis1000/asmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1

sourceHugging Faceapache-2.0updated 4mo agoView on Hugging Face
0likes105downloads
Model Card

asmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1

GGUF quantizations of a fine-tuned model for translating Japanese ASMR transcriptions (ASR/Whisper output) into Traditional Chinese.

The model normalizes imperfect audio transcriptions, applies domain-specific glossaries, and translates character dialogue while retaining emotion and nuances.

This variant preserves the upstream Qwen3.5 MTP / speculative-decoding head in GGUF format so it can be used with MTP-capable llama.cpp builds.

Standard Mode

The traditional output format where only the translated text is returned.

Available Quantizations

QuantizationFilenameSizeDescription
q4kmasmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1-q4_k_m.gguf5.4 GBGood balance of quality and size
q6_kasmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1-q6_k.gguf7.0 GBHigher quality, moderate size
q8_0asmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1-q8_0.gguf9.1 GBNear-lossless quality
bf16asmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1-bf16.gguf17.1 GBFull BF16, no quantization loss

Prompt Example

`text
將以下日語ASMR逐字稿翻譯成繁體中文。

音軌:track01_示例音軌
場景說明:主角與青梅竹馬在校園下午的對話...

術語表(請嚴格使用zh欄位的譯名):
{
  "cvs": [],
  "characters": [],
  "terms": [{"ja": "放課後", "zh": "放學後"}]
}

翻譯前請靜默修正下列Whisper識別錯誤:
- 重複片語(連續3次以上且無變化):僅保留一次
- 錯字/同音異字:依上下文修正
- 字幕版權行(字幕:/翻訳:/QQ/LINE水印):text設為null
- 錯誤專有名詞:依術語表修正

翻譯規則:
- 呻吟與氣息聲(あ、ん、はあ)→ 自然對應(啊、嗯、哈、呼)
- 擬聲詞:日語形式翻譯(パンパン→啪啪);中文形式保留原樣
- 保留角色語氣與口吻
- text欄位只輸出譯文,不加注釋或括號說明

輸入:逐字稿JSON陣列 — {"id": <n>, "text": "<日文>", "start": <ms>, "end": <ms>}

輸出:將連續構成同一句話的片段合併,JSON陣列格式:
{"ids": [<n>, ...], "text": "<繁體中文>", "start": <最早ms>, "end": <最晚ms>}

字幕版權行:{"ids": [<n>], "text": null, "start": <ms>, "end": <ms>}
每個輸入id必須恰好出現在一個輸出項中。

逐字稿:
[
  {"id": 1, "text": "ねぇ、放課後、", "start": 3000, "end": 5000},
  {"id": 2, "text": "一緒に帰らない?", "start": 5000, "end": 7000}
]

Example Output:

json
[{"ids": [1, 2], "text": "欸,放學後,要不要一起回去?", "start": 3000, "end": 7000}]

Usage

llama-server

bash
llama-server -m asmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1-q4_k_m.gguf -c 4096 --port 8080

llama-cli

bash
llama-cli -m asmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1-q4_k_m.gguf -p "<your prompt>" -n 2048

llama-cli with MTP

bash
llama-cli -m asmr-qwen3.5-9b-zh-tw-mtp-gguf-v0.1-q4_k_m.gguf --spec-type draft-mtp --spec-draft-n-max 6 -p "<your prompt>" -n 2048

Structured Decoding (Recommended)

This model outputs JSON arrays. Using structured decoding (e.g. GBNF grammar or JSON schema constraints) avoids wasted computation on malformed output and guarantees valid JSON on every generation.

JSON Schema:

json
{
  "type": "array",
  "items": {
    "type": "object",
    "properties": {
      "ids": {
        "type": "array",
        "items": {
          "type": "integer"
        },
        "minItems": 1
      },
      "text": {
        "anyOf": [
          {
            "type": "string"
          },
          {
            "type": "null"
          }
        ]
      },
      "start": {
        "type": "integer"
      },
      "end": {
        "type": "integer"
      }
    },
    "required": [
      "ids",
      "text",
      "start",
      "end"
    ],
    "additionalProperties": false
  },
  "minItems": 1
}

Supported by llama.cpp (--json-schema), vLLM, and outlines.

Training Details

  • —Base model: unsloth/Qwen3.5-9B
  • —Method: LoRA (r=16, alpha=16)
  • —Target modules: qproj, kproj, vproj, upproj, gateproj, downproj, o_proj
  • —Locale: zh-tw (Traditional Chinese)
  • —Mode: Standard Mode
  • —Max sequence length: 4096
  • —Precision: bf16