Automatic Speech Recognition
MLX
ONNX
GGUF
Rust
English
Chinese
audio8
streaming-asr
quantized
experimental
Instructions to use Reza2kn/Audio8-ASR-Infinite-Compressed with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use Reza2kn/Audio8-ASR-Infinite-Compressed with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download Reza2kn/Audio8-ASR-Infinite-Compressed --local-dir Audio8-ASR-Infinite-Compressed
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
File size: 4,912 Bytes
8bdbdfc | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 | {
"status": "complete",
"summary": {
"kind": "caption_stream_complete",
"status": "complete",
"audio_samples": 93680,
"audio_seconds": 5.855,
"segments": 1,
"selected_samples": 90864,
"unconsumed_selected_samples": 0,
"emissions": 85,
"text_bytes": 88,
"model_seconds": 4.573718884000001,
"reset_seconds": 8.12e-07,
"vad_seconds": 0.053173863999999973,
"read_seconds": 1.9281761100000006,
"sleep_seconds": 0,
"elapsed_seconds": 6.558700714,
"work_rtf": 0.7902465516652434,
"peak_cache_bytes": 210169856,
"max_retained_input_samples": 6912,
"input_sha256": "22d472b40f913206b6917115d136306be88df3b278e77f68be5910423250fa4d",
"generated_ids_sha256": "4cab5d89ace01105314a3ff9ea386dd217c2cb319c74d0295a34ead46d3c44f7",
"all_source_audio_observed": true,
"measured_word_aligned_latency": false,
"production_accuracy_approved": false
},
"score": {
"case_id": "1272-128104-0000",
"language": "en",
"input_sha256": "22d472b40f913206b6917115d136306be88df3b278e77f68be5910423250fa4d",
"raw_hypothesis": "Mr. Quilter is the apostle of the middle classes, and we are glad to welcome his gospel.",
"normalized_reference": "mister quilter is the apostle of the middle classes and we are glad to welcome his gospel",
"normalized_hypothesis": "mr quilter is the apostle of the middle classes and we are glad to welcome his gospel",
"metric": "WER",
"counts": {
"hits": 16,
"substitutions": 1,
"deletions": 0,
"insertions": 0,
"reference_units": 17,
"hypothesis_units": 17,
"errors": 1,
"error_rate": 0.058823529411764705
},
"longest_contiguous_deleted_reference_units": 0,
"aligned_word_timing": {
"measured": false,
"actual_arrival_paced": true,
"timing_basis": "Independent host receipt of flushed JSONL since paced20ms producer start; all source words retained",
"correctly_matched_units": 0,
"ambiguous_matched_units": 0,
"ambiguity_method": "all minimum unit-cost Levenshtein paths, without timing-based selection",
"unambiguous_matched_units": 0,
"unambiguous_maximum_seconds": null,
"unmatched_reference_units": 1,
"matched_reference_fraction": null,
"min_seconds": null,
"p50_seconds": null,
"p95_seconds": null,
"maximum_seconds": null,
"matched_units_above_one_second": null,
"all_matched_units_within_one_second": null,
"whole_release_pass": false,
"limitation": "Lag summaries use one deterministic text alignment and are descriptive when alternative optimal alignments exist. Only unambiguous matches can support per-word timing claims. Omissions/substitutions remain errors, never zero-latency successes. Published reference timing uncertainty and overlapping-speaker word order are not independently corrected."
},
"alignment": [],
"error_spans": [
{
"type": "substitute",
"reference_range": [
0,
1
],
"hypothesis_range": [
0,
1
]
}
],
"first_nonempty_delta_seconds": 1.3183811120688915,
"production_accuracy_approved": false
},
"segments": [
{
"id": 0,
"start_sample": 2816,
"decision_sample": 9216,
"end_sample": 93680,
"final_eof": true,
"windows": 85,
"unconsumed_samples": 0
}
],
"all_ids_exact_expected": null,
"independent_tokenizer_oracle": true,
"max_queue_seconds": 0.654069003,
"no_synthetic_nonfinal_eof": true,
"retention_bounded": true,
"release_qualification": false,
"independent_delta_token_clock_oracle": true,
"feeder": {
"chunks": 293,
"samples": 93680,
"duration_seconds": 5.8550937010440975,
"max_write_seconds": 5.0443923100829124e-05,
"max_delivery_lateness_seconds": 0.0013744130358099849,
"process_started_monotonic": 1189174.972518553,
"feeder_started_monotonic": 1189177.764669274,
"ready_received_monotonic": 1189177.764503835,
"clock_scope": "Feeder starts after host receipt of native readiness. Native and observer clocks are separately recorded; no microphone capture clock."
},
"command": [
"/bin/sh",
"/dev/shm/audio8-q3-public-linux-consumer-v1/run-cpu-linux-x86_64.sh",
"-",
"en",
"/dev/shm/audio8-q3-public-linux-consumer-v1/verification/en.a8pfx"
],
"invoked_shipped_launcher": true,
"memory": {
"os_peak_process_rss_bytes": 2025762816,
"os_peak_memory_footprint_bytes": null,
"footprint_available": false,
"rss_source_unit": "KiB (1024 bytes)",
"elapsed_seconds": 9.4,
"user_seconds": 26.12,
"system_seconds": 2.63,
"scope": "Fresh native child process including loading and inference; excludes Python harness. RSS and footprint are separate measurements and must not be summed."
}
}
|