Add ASR pipeline metadata

.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tekken.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,84 @@
+---
+license: apache-2.0
+language:
+- en
+- fr
+- de
+- es
+- it
+- pt
+- nl
+- hi
+base_model: mistralai/Voxtral-Mini-3B-2507
+library_name: mlx
+pipeline_tag: automatic-speech-recognition
+tags:
+- mlx
+- speech-recognition
+- automatic-speech-recognition
+---
+
+# Voxtral Mini 3B 2507 — MLX 8-bit affine
+
+MLX weights for [Voxtral Mini 3B 2507](https://huggingface.co/mistralai/Voxtral-Mini-3B-2507),
+ready for local speech-to-text with
+[speech-swift](https://github.com/soniqo/speech-swift).
+
+The FP16, INT5, and INT8 family passed the isolated WER, RTF, throughput, RSS, and physical-footprint compatibility gate on 2026-07-22 using English FLEURS.
+
+| Parameters | Precision | Sample rate | Languages |
+|---:|---:|---:|---:|
+| 3B | 8-bit affine | 16 kHz | 8 |
+
+## Files
+
+| File | Description |
+|---|---|
+| `model*.safetensors` | MLX model weights |
+| `config.json` and `params.json` | Model and runtime configuration |
+| `tekken.json` | Tekken tokenizer data |
+
+## Usage
+
+```bash
+speech transcribe recording.wav --engine voxtral --model int8
+```
+
+```swift
+import VoxtralASR
+
+let model = try await VoxtralModel.load(
+    "aufklarer/Voxtral-Mini-3B-2507-MLX-8bit"
+)
+let text = model.transcribe(
+    audio: samples,
+    sampleRate: sourceSampleRate,
+    language: "en"
+)
+```
+
+## Benchmark
+
+English FLEURS read speech, 80 utterances (759.56 seconds), Apple M5 Pro:
+
+| Variant | Bundle | WER | Mean RTF | Peak RSS |
+|---|---:|---:|---:|---:|
+| FP16 | 8.71 GiB | 4.633% | 0.1305 | 9,091 MiB |
+| INT5 | 3.77 GiB | 4.744% | 0.0739 | 4,035 MiB |
+| INT8 | 5.18 GiB | 4.578% | 0.0906 | 5,487 MiB |
+
+These read-speech measurements are not Artificial Analysis Conversational or
+ElevenLabs Scribe results.
+
+## Source and license
+
+The base model is [mistralai/Voxtral-Mini-3B-2507](https://huggingface.co/mistralai/Voxtral-Mini-3B-2507) and is
+released under Apache 2.0.
+
+## Links
+
+- [Voxtral guide](https://soniqo.audio/guides/voxtral)
+- [speech-swift](https://github.com/soniqo/speech-swift)
+- [Soniqo documentation](https://soniqo.audio/getting-started)
+- [Soniqo](https://soniqo.audio)
+- [Blog](https://soniqo.audio/blog)

config.json

@@ -0,0 +1,63 @@
+{
+    "architectures": [
+        "VoxtralForConditionalGeneration"
+    ],
+    "audio_config": {
+        "activation_dropout": 0.0,
+        "activation_function": "gelu",
+        "attention_dropout": 0.0,
+        "dropout": 0.0,
+        "head_dim": 64,
+        "hidden_size": 1280,
+        "initializer_range": 0.02,
+        "intermediate_size": 5120,
+        "layerdrop": 0.0,
+        "max_source_positions": 1500,
+        "model_type": "voxtral_encoder",
+        "num_attention_heads": 20,
+        "num_hidden_layers": 32,
+        "num_key_value_heads": 20,
+        "num_mel_bins": 128,
+        "scale_embedding": false,
+        "vocab_size": 51866
+    },
+    "audio_token_id": 24,
+    "hidden_size": 3072,
+    "model_type": "voxtral",
+    "projector_hidden_act": "gelu",
+    "quantization": {
+        "group_size": 64,
+        "bits": 8,
+        "mode": "affine"
+    },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 8,
+        "mode": "affine"
+    },
+    "text_config": {
+        "attention_bias": false,
+        "attention_dropout": 0.0,
+        "head_dim": 128,
+        "hidden_act": "silu",
+        "hidden_size": 3072,
+        "initializer_range": 0.02,
+        "intermediate_size": 8192,
+        "max_position_embeddings": 131072,
+        "mlp_bias": false,
+        "model_type": "llama",
+        "num_attention_heads": 32,
+        "num_hidden_layers": 30,
+        "num_key_value_heads": 8,
+        "pretraining_tp": 1,
+        "rms_norm_eps": 1e-05,
+        "rope_scaling": null,
+        "rope_theta": 100000000.0,
+        "sliding_window": null,
+        "use_cache": true,
+        "vocab_size": 131072
+    },
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.54.0.dev0",
+    "vocab_size": 131072
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 11,
+  "transformers_version": "4.54.0.dev0"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d497498404c4717e4be6f04b65ed5ab88c17ea08a4671c21dcf00f0f8a2d0f54
+size 5368741916

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdf722bf98262d6e310eb5ef92e562b415d557dc037b4e317c1b096c534b64fc
+size 197276087

params.json

@@ -0,0 +1,34 @@
+{
+  "dim": 3072,
+  "n_layers": 30,
+  "head_dim": 128,
+  "hidden_dim": 8192,
+  "n_heads": 32,
+  "n_kv_heads": 8,
+  "rope_theta": 100000000.0,
+  "norm_eps": 1e-05,
+  "vocab_size": 131072,
+  "max_position_embeddings": 32768,
+  "multimodal": {
+    "whisper_model_args": {
+      "encoder_args": {
+        "dim": 1280,
+        "n_layers": 32,
+        "head_dim": 64,
+        "hidden_dim": 5120,
+        "n_heads": 20,
+        "vocab_size": 51866,
+        "max_source_positions": 1500,
+        "audio_encoding_args": {
+          "sampling_rate": 16000,
+          "num_mel_bins": 128,
+          "hop_length": 160,
+          "window_size": 400
+        }
+      },
+      "downsample_args": {
+        "downsample_factor": 4
+      }
+    }
+  }
+}

preprocessor_config.json

@@ -0,0 +1,15 @@
+{
+  "chunk_length": 30,
+  "dither": 0.0,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 128,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "VoxtralProcessor",
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

tekken.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4aaf3836c2a5332f029ce85a7a62255c966f47b6797ef81dedd0ade9c862e4a8
+size 14894206