Mohammad Hijjawi b0fa25b12c fix: keep token separators in timestamp_sentence raw_text (#3746) ...

timestamp_sentence() stripped the trailing space from sentence_text_seg
on every token instead of once when the sentence is emitted, so the
per-sentence raw_text lost all separators ("iphone pro max" became
"iphonepromax"). Trim it only at the sentence boundary, as
timestamp_sentence_en() already does.

2026-10-01 12:48:39 +08:00

..

run_test.py

fix: repair KWS symbol cleanup regex

2026-05-21 13:05:51 +00:00

test_amp_compat.py

fix: use torch.amp via compatibility shim (#3518)

2026-08-26 20:10:53 +08:00

test_amp_device_type.py

fix(amp): resolve autocast/GradScaler device type for non-CUDA accelerators (#3735)

2026-09-30 16:48:49 +08:00

test_api_signatures.py

docs: preserve API signatures and discover AutoModelVLLM

2026-09-07 03:38:20 +00:00

test_asr_inference_pipeline.py

Dev gzf exp (#1654)

2024-04-24 16:03:38 +08:00

test_asr_vad_punc_inference_pipeline.py

Dev gzf exp (#1654)

2024-04-24 16:03:38 +08:00

test_auto_model_ncpu.py

fix(campplus): bound clustering thread usage and spectrum work (#3699)

2026-09-19 10:32:10 +08:00

test_auto_model.py

fix(auto-model): report vda model typo (#3443)

2026-07-30 04:13:35 +08:00

test_campplus_utils.py

fix: count speaker overlap once per diarization segment

2026-09-28 12:50:06 +00:00

test_canonical_qwenaudio_links.py

Use QwenAudio canonical repository links (#3412)

2026-07-24 19:01:36 +08:00

test_check_funasr_website_static.py

fix(website): correct LiteLLM transcription route and guards

2026-09-29 03:10:14 +00:00

test_checkpoint_metric_distributed.py

fix(training): rank checkpoints only with valid validation metrics

2026-09-08 21:07:19 +00:00

test_checkpoint_metric_presence.py

fix(training): rank checkpoints only with valid validation metrics

2026-09-08 21:07:19 +00:00

test_cli_docs_contract.py

docs: align bilingual CLI guide with parser and output contracts

2026-09-06 22:37:12 +00:00

test_cli.py

fix(cli): preserve readable subtitle boundaries

2026-08-30 19:57:12 +00:00

test_cluster_backend.py

fix(diarization): scale clustering with known speaker counts

2026-08-20 01:15:55 +08:00

test_collect_growth_metrics.py

ops(growth): track SGLang Fun-ASR CI gate (#3510)

2026-08-17 09:44:30 +08:00

test_collect_website_traffic.py

Track privacy-safe website traffic (#3392)

2026-07-23 13:00:02 +08:00

test_community_projects_entries.py

docs: surface merged speech-to-speech SenseVoice integration

2026-09-30 16:36:43 +00:00

test_ct_transformer_docs.py

test(punctuation): guard SenseVoice documentation on exact PR heads

2026-09-15 14:30:57 +00:00

test_data2vec_compute_var.py

fix: use y.device instead of hardcoded .cuda() in data2vec compute_var (#3710)

2026-09-21 10:28:58 +08:00

test_docs_funasr_install_commands.py

test(docs): derive current release checks from package version

2026-09-09 03:42:38 +00:00

test_dynamic_streaming_vad.py

fix(vad): exclude idle time from dynamic silence schedule

2026-09-08 20:15:17 +00:00

test_eend_feature_compatibility.py

fix: support NumPy 2 with verified scientific-stack compatibility (#3674)

2026-09-09 10:59:37 +08:00

test_fsmn_vad_dynamic_silence.py

fix: respect explicit VAD silence threshold

2026-06-11 15:16:24 +00:00

test_fsmn_vad_streaming_buffers.py

fix: bound streaming VAD frame buffers (#3236)

2026-07-17 17:24:23 +08:00

test_fun_asr_nano_autocast_device.py

Fix Fun-ASR-Nano NPU autocast device selection (#3113)

2026-07-07 17:45:00 +08:00

test_fun_asr_nano_byte_fragment_timestamps.py

fix(nano): correct punctuation timestamps across VAD merges (#3703)

2026-09-25 10:16:47 +08:00

test_fun_asr_nano_ctc_batch_fallback.py

Fix Fun-ASR-Nano CTC batch fallback (#3187)

2026-07-08 15:07:02 +08:00

test_fun_asr_nano_lora_injection.py

test(nano): make LoRA tests deterministic (#3459)

2026-08-03 15:24:08 +08:00

test_fun_asr_nano_missing_ctc_weights.py

fix: disable timestamps when CTC weights are missing (#3211)

2026-07-13 23:57:17 +08:00

test_fun_asr_nano_openai_response.py

fix: limit vLLM VAD segments to 30 seconds (#3420)

2026-07-26 14:53:00 +08:00

test_fun_asr_nano_punctuation_timestamps.py

fix(nano): correct punctuation timestamps across VAD merges (#3703)

2026-09-25 10:16:47 +08:00

test_fun_asr_nano_repetition_penalty.py

fix: avoid CUDA crash from repetition_penalty in Fun-ASR-Nano vLLM prompt-embeds mode (#2948) (#2974)

2026-06-15 09:31:04 -07:00

test_fun_asr_nano_semantic_symbol_timestamps.py

fix(nano): correct punctuation timestamps across VAD merges (#3703)

2026-09-25 10:16:47 +08:00

test_fun_asr_nano_train_opts.py

[perf] Fun-ASR-Nano fine-tuning on B200: 0.244 s to 0.068 s per step at 1 GPU, mostly from cuDNN SDPA plan builds on every new batch shape (#3705)

2026-09-21 10:20:05 +08:00

test_fun_asr_nano_vllm_dtype.py

fix(vllm): honor fp32 decoder fallback

2026-09-01 04:57:37 +00:00

test_fun_asr_nano_vllm_prompt.py

fix(fun-asr-nano): normalize ISO language hints

2026-09-02 03:44:23 +00:00

test_funasr_onnx_installed.py

release(onnx): prepare 0.4.3 with installed artifact regressions

2026-09-09 22:58:58 +00:00

test_funasr_onnx_release.py

release(onnx): prepare 0.4.3 with installed artifact regressions

2026-09-09 22:58:58 +00:00

test_generate_subtitle.py

fix(subtitle): preserve more VAD context by default

2026-09-06 09:28:10 +00:00

test_generated_api_docs.py

docs: preserve API signatures and discover AutoModelVLLM

2026-09-07 03:38:20 +00:00

test_gh_pages_moss_publication.py

docs: publish MOSS discovery on GitHub Pages

2026-09-02 16:41:09 +00:00

test_glm_asr_vllm_dedup_keys.py

fix(glm_asr): dedup result keys for duplicate audio basenames (#3014)

2026-06-22 11:40:12 +08:00

test_glm_asr_vllm_dtype.py

fix(glm_asr): warn when vLLM dtype=fp16 (degraded output) (#2993)

2026-06-20 22:40:03 +08:00

test_glm_asr_vllm_sampling.py

fix(glm_asr): honor sampling params in vLLM generate() (#2997)

2026-06-20 22:40:06 +08:00

test_gradio_app.py

feat: add explicit MOSS serving profiles to Gradio client

2026-09-07 19:12:22 +00:00

test_growth_plan_doc.py

ops(growth): track SGLang Fun-ASR CI gate (#3510)

2026-08-17 09:44:30 +08:00

test_growth_target_doc.py

docs: correct strict ecosystem growth target

2026-09-15 23:34:51 +00:00

test_hf_model_aliases.py

fix: route paraformer-en to the English HF model (#3466)

2026-08-04 10:06:33 +08:00

test_kws_docs.py

docs(kws): add bilingual task guides and tested session examples

2026-09-07 08:13:07 +00:00

test_kws_optional_output.py

fix(kws): honor optional output directory on every call

2026-09-07 07:01:26 +00:00

test_kws_streaming_continuity.py

fix(kws): preserve streaming frames across packet and EOS boundaries

2026-09-07 07:39:13 +00:00

test_kws_utils.py

fix: repair KWS symbol cleanup regex

2026-05-21 13:05:51 +00:00

test_learning_docs_contract.py

docs: rebuild the learning, model and serving documentation system

2026-09-06 14:11:58 +00:00

test_lm_pipeline.py

Dev gzf exp (#1654)

2024-04-24 16:03:38 +08:00

test_load_audio_bytes.py

Decode container-formatted audio bytes before inference (#3435)

2026-07-27 14:55:49 +00:00

test_markdown_relative_links.py

docs: fix broken markdown links

2026-07-22 22:02:41 +08:00

test_mcp_server_example.py

Document Glama MCP submission checklist (#3084)

2026-06-30 17:25:45 +08:00

test_metrics_common_rapidfuzz.py

Migrate editdistance metrics to rapidfuzz with empty-reference guard

2026-07-07 15:00:51 +08:00

test_moss_transcribe_diarize_docs.py

docs: add MOSS to model zoo indexes (#3636)

2026-09-03 17:15:22 +08:00

test_moss_transcribe_diarize_model.py

feat(moss): add SGLang AutoModel backend

2026-08-30 15:21:25 +00:00

test_numpy_cluster_compatibility.py

fix: support NumPy 2 with verified scientific-stack compatibility (#3674)

2026-09-09 10:59:37 +08:00

test_numpy_compatibility.py

fix: support NumPy 2 with verified scientific-stack compatibility (#3674)

2026-09-09 10:59:37 +08:00

test_online_cpu_dockerfile.py

ci: cache online CPU Docker layers

2026-08-30 18:42:36 +00:00

test_openai_api_docker_validation.py

Add OpenAI API Docker validation helper (#3371)

2026-07-23 06:49:57 +08:00

test_openai_api_n8n_compat.py

fix(openai-api): support n8n audio alias in example

2026-09-02 03:24:25 +00:00

test_paraformer_timestamp_contract.py

fix(paraformer): correct timestamp alpha and peak argument order (#3716)

2026-09-19 10:42:09 +08:00

test_paraformer_timestamp_padding.py

fix(paraformer): correct timestamp alpha and peak argument order (#3716)

2026-09-19 10:42:09 +08:00

test_postprocess_hotwords.py

fix(hotwords): stop fuzzy postprocess matching from rewriting nearby text (#3744)

2026-10-01 12:47:56 +08:00

test_punc_model_none.py

fix: honor submodel device placement across inference calls

2026-09-14 09:20:40 +08:00

test_punctuation_pipeline.py

Dev gzf exp (#1654)

2024-04-24 16:03:38 +08:00

test_pypi_metadata.py

fix: support NumPy 2 with verified scientific-stack compatibility (#3674)

2026-09-09 10:59:37 +08:00

test_python_api_docs_contract.py

fix: honor submodel device placement across inference calls

2026-09-14 09:20:40 +08:00

test_qwen3_asr_dependency_check.py

Add Qwen3-ASR dependency guard

2026-06-29 22:52:44 +08:00

test_qwen3_asr_vllm_offline_example.py

feat(examples): add MOSS diarization to offline vLLM sample

2026-09-08 22:11:20 +00:00

test_qwen3_asr_ws_example.py

feat: add Qwen3-ASR streaming WebSocket server with notes (#3035)

2026-06-28 20:29:47 +08:00

test_readme_deployment_hub_links.py

docs: add a concise community discovery entrypoint (#3617)

2026-09-02 17:37:23 +08:00

test_readme_task_entrypoints.py

docs: distinguish native Transformers model downloads

2026-09-10 06:39:10 +00:00

test_realtime_ws_benchmark.py

feat(benchmark): report first nonempty transcript latency

2026-09-28 13:36:31 +00:00

test_realtime_ws_service.py

fix(realtime): reduce default partial decode window (#3632)

2026-09-03 16:32:40 +08:00

test_reference_docs_contract.py

docs: finish reference navigation and correct model and vLLM contracts

2026-09-06 17:43:37 +00:00

test_release_package_data.py

fix: include runtime package data (#3445)

2026-07-30 05:32:44 +08:00

test_release_runtime_downloads.py

ci(release): attach Python packages automatically

2026-08-30 20:47:50 +00:00

test_release_version_contract.py

Release FunASR 1.4.16 (#3719)

2026-09-18 12:03:34 +08:00

test_repository_roadmap_docs.py

docs: align bilingual roadmap with delivered milestones

2026-09-15 17:47:50 +00:00

test_runtime_sdk_download_tool.py

fix(download): preserve snapshot errors (#3565)

2026-08-30 02:36:45 +08:00

test_sensevoice_continual_finetuning_docs.py

fix(training): rank checkpoints only with valid validation metrics

2026-09-08 21:07:19 +00:00

test_sensevoice_tensorrt_builder.py

feat(runtime): add SenseVoice TensorRT deployment (#3463)

2026-08-04 02:22:57 +08:00

test_sensevoice_tokenizer_special_tokens.py

fix(tokenizer): treat special-token strings as text to prevent batch crashes (#3110)

2026-07-07 14:42:36 +08:00

test_sensevoice_word_timestamps.py

fix(sensevoice): keep words separate at standalone boundaries

2026-09-29 06:12:59 +00:00

test_server_app_openai_segments.py

feat(server): support n8n OpenAI audio transcription

2026-09-02 03:20:13 +00:00

test_service_docs_contract.py

docs: align deployment matrix recipes and fix copy icon contrast

2026-09-07 22:10:59 +00:00

test_silero_vad_adapter.py

fix(vad): preserve Silero ONNX CPU inputs

2026-08-14 07:35:21 +00:00

test_source_archive_portability.py

fix: keep source archives portable

2026-09-03 22:08:14 +08:00

test_speaker_emotion_docs.py

docs: add speaker and emotion task guides

2026-09-07 08:53:08 +00:00

test_sphinx_links.py

docs: consolidate Chinese vLLM guides with checked compatibility mirror

2026-09-07 10:21:57 +00:00

test_streaming_vad_docs.py

docs: add bilingual streaming VAD task guide

2026-09-07 05:53:40 +00:00

test_submodel_device.py

fix(auto): warn when unavailable accelerators fall back to CPU

2026-09-30 12:02:57 +00:00

test_sv_inference_pipeline.py

Dev gzf exp (#1654)

2024-04-24 16:03:38 +08:00

test_text_normalization_punctuation.py

fix: handle Unicode punctuation post-processing (#3238)

2026-07-15 21:03:59 +08:00

test_timestamp_tools.py

fix: keep token separators in timestamp_sentence raw_text (#3746)

2026-10-01 12:48:39 +08:00

test_top_level_import.py

fix: lazy load top-level AutoModel exports

2026-07-19 18:27:49 +08:00

test_torchaudio_optional.py

Remove hard torchaudio dependency for inference; add kaldi-native-fbank fbank backend (#3526)

2026-08-27 00:23:11 +08:00

test_tp_pipeline.py

Dev gzf exp (#1654)

2024-04-24 16:03:38 +08:00

test_train_ds_distributed_config.py

fix(train): initialize DeepSpeed mode from resolved config (#3471)

2026-08-05 11:35:53 +08:00

test_trainer_ds_unvalidated_ckpt.py

fix(trainer): avoid KeyError when a checkpoint is saved at an unvalidated step (#3455)

2026-08-03 14:19:25 +08:00

test_training_docs_contract.py

docs: rebuild the learning, model and serving documentation system

2026-09-06 14:11:58 +00:00

test_vad_inference_pipeline.py

Dev gzf exp (#1654)

2024-04-24 16:03:38 +08:00

test_vehicle_plate_finetuning_guide.py

docs: add vehicle plate fine-tuning guide (#3616)

2026-09-02 16:35:58 +08:00

test_vllm_examples_contract.py

docs: finish reference navigation and correct model and vLLM contracts

2026-09-06 17:43:37 +00:00

test_vllm_guide_sync_docs.py

docs: consolidate Chinese vLLM guides with checked compatibility mirror

2026-09-07 10:21:57 +00:00

test_vllm_model_source_docs.py

test: keep rendering checks in the product-site suite

2026-09-08 18:41:03 +00:00

test_websocket_file_finalization.py

fix(websocket): finalize file input on end control (#3440)

2026-07-29 20:24:01 +08:00