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
Download evaluation/exposed-native-stream.json from Reza2kn/Audio8-ASR-Infinite-Compressed: direct link, hf CLI and curl.
- Browser
- Download file 2.63 kB
-
https://huggingface.co/Reza2kn/Audio8-ASR-Infinite-Compressed/resolve/main/evaluation/exposed-native-stream.json
- Command line
-
hf download hf://Reza2kn/Audio8-ASR-Infinite-Compressed/evaluation/exposed-native-stream.json
-
curl -L -o exposed-native-stream.json https://huggingface.co/Reza2kn/Audio8-ASR-Infinite-Compressed/resolve/main/evaluation/exposed-native-stream.json
2.63 kB
| { | |
| "status": "verified_exposed_diagnostic", | |
| "production_ready": false, | |
| "model_changed": true, | |
| "counts": { | |
| "hits": 302, | |
| "substitutions": 39, | |
| "deletions": 137, | |
| "insertions": 4, | |
| "reference_units": 478, | |
| "hypothesis_units": 345, | |
| "errors": 180, | |
| "error_rate": 0.37656903765690375 | |
| }, | |
| "prior_q4_counts": { | |
| "hits": 298, | |
| "substitutions": 46, | |
| "deletions": 134, | |
| "insertions": 5, | |
| "reference_units": 478, | |
| "hypothesis_units": 349, | |
| "errors": 185, | |
| "error_rate": 0.38702928870292885 | |
| }, | |
| "word_timing": { | |
| "measured": true, | |
| "actual_arrival_paced": true, | |
| "timing_basis": "Independent host receipt of flushed JSONL since paced20ms producer start; all source words retained", | |
| "correctly_matched_units": 302, | |
| "ambiguous_matched_units": 9, | |
| "ambiguity_method": "all minimum unit-cost Levenshtein paths, without timing-based selection", | |
| "unambiguous_matched_units": 293, | |
| "unambiguous_maximum_seconds": 5.347705096863194, | |
| "unmatched_reference_units": 176, | |
| "matched_reference_fraction": 0.6317991631799164, | |
| "min_seconds": -2.329683994008221, | |
| "p50_seconds": 0.4744840469956415, | |
| "p95_seconds": 0.613818718912114, | |
| "maximum_seconds": 5.347705096863194, | |
| "matched_units_above_one_second": 1, | |
| "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." | |
| }, | |
| "independent_delta_token_clock_oracle": true, | |
| "all_ids_logits_clocks_exact_q3_v1": true, | |
| "memory": { | |
| "os_peak_process_rss_bytes": 2126622720, | |
| "os_peak_memory_footprint_bytes": null, | |
| "footprint_available": false, | |
| "rss_source_unit": "KiB (1024 bytes)", | |
| "elapsed_seconds": 324.52, | |
| "user_seconds": 887.31, | |
| "system_seconds": 73.95, | |
| "scope": "Fresh native child process including loading and inference; excludes Python harness. RSS and footprint are separate measurements and must not be summed." | |
| }, | |
| "mature_work_rtf_including_trims": 0.9405229254891304, | |
| "peak_non_eof_receipt_backlog_seconds": 0.3471983608733922, | |
| "raw_sha256": "cc977ded1479cbccd64494a800a5ed43570cba91d71bce851da4919dcf7ed0bf", | |
| "scope": "Previously exposed full meeting; memory, sustained CPU work and caption-after-word clocks. No protected evaluation acceptance." | |
| } | |