{ "high_confidence_ranking": [ "volcengine-current-asr", "qwen3-asr-1.7b-plus-external-forced-aligner", "moss-transcribe-diarize-0.9b" ], "item_count": 23, "items": [ { "MOSS_candidate": { "both_within_1s_of_machine_reference": false, "end": 179.66, "start": 174.43 }, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 179.474, "start": 177.634 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 179.48, "start": 177.8 }, "audio_clip": { "duration_seconds": 5.72, "end_seconds_in_window": 181.52, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-01-zhang-front.wav", "sha256": "a409bb8ef5209433951a3a0ae1a01402252a70c4b935c255381596d5384e560b", "start_seconds_in_window": 175.8 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 179.52, "start": 177.8 }, "priority": "P0", "reference_segment_index": 33, "review_id": "timestamp-decisive-01", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_both_within_1s_outcome", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "这是，Is it real？", "window_id": "front" }, { "MOSS_candidate": { "both_within_1s_of_machine_reference": false, "end": 186.35, "start": 185.48 }, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": false, "end": 186.354, "start": 185.474 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 187.56, "start": 185.6 }, "audio_clip": { "duration_seconds": 5.96, "end_seconds_in_window": 189.56, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-02-zhang-front.wav", "sha256": "a38fbfe11b9e2fd040bc43481261fe6c27551e7fcc4c6d5e519c9b7ae9a325a9", "start_seconds_in_window": 183.6 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 187.56, "start": 185.6 }, "priority": "P0", "reference_segment_index": 35, "review_id": "timestamp-decisive-02", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_both_within_1s_outcome", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "就难以置信。", "window_id": "front" }, { "MOSS_candidate": { "both_within_1s_of_machine_reference": true, "end": 250.2, "start": 239.84 }, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 250.046, "start": 239.886 }, "Volc_candidate": { "both_within_1s_of_machine_reference": false, "end": 255.43, "start": 239.87 }, "audio_clip": { "duration_seconds": 14.48, "end_seconds_in_window": 252.35, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-03-zhang-front.wav", "sha256": "5755780f783ecd944b81a96207ac55d5d3fd684ed612fd070a6c1d4b9e5112e8", "start_seconds_in_window": 237.87 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 250.35, "start": 239.87 }, "priority": "P0", "reference_segment_index": 48, "review_id": "timestamp-decisive-03", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_both_within_1s_outcome", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "我们内部核心开了一个会，当然我其实已经那段最难受的记忆，我已经有点泛白了哈，真的从9万多人减到剩1000人以内的时候，那段我确实记忆有点泛白了。", "window_id": "front" }, { "MOSS_candidate": { "both_within_1s_of_machine_reference": true, "end": 262.82, "start": 250.42 }, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 262.676, "start": 250.446 }, "Volc_candidate": { "both_within_1s_of_machine_reference": false, "end": 262.809, "start": 239.87 }, "audio_clip": { "duration_seconds": 16.32, "end_seconds_in_window": 264.79, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-04-zhang-front.wav", "sha256": "3dc4392b77c82f88127b3a3ced12129af8032098b0108d1756fef5954da76abf", "start_seconds_in_window": 248.47 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 262.79, "start": 250.47 }, "priority": "P0", "reference_segment_index": 49, "review_id": "timestamp-decisive-04", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_both_within_1s_outcome", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "但是我我我记得当时就其实这个体量的减员是非常难受的，当然对我来说哈，首先第一难受的是你很多兄弟嘛，你要 say goodbye 然后我还印象深刻，有些 hr 我们原来在全国五十几个城市都有分公司。", "window_id": "front" }, { "MOSS_candidate": { "both_within_1s_of_machine_reference": false, "end": 363.85, "start": 359.92 }, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 363.684, "start": 362.484 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 363.89, "start": 362.53 }, "audio_clip": { "duration_seconds": 5.36, "end_seconds_in_window": 365.89, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-05-zhang-front.wav", "sha256": "52549aafab8cf0bd5512c3758735ec918aa7ce8edb611512785bfc98e89e4a44", "start_seconds_in_window": 360.53 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 363.89, "start": 362.53 }, "priority": "P0", "reference_segment_index": 78, "review_id": "timestamp-decisive-05", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_both_within_1s_outcome", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "啊，所以最后才做了决定。", "window_id": "front" }, { "MOSS_candidate": { "both_within_1s_of_machine_reference": false, "end": 456.01, "start": 453.58 }, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 453.852, "start": 453.612 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 454.02, "start": 453.74 }, "audio_clip": { "duration_seconds": 4.28, "end_seconds_in_window": 456.02, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-06-zhang-front.wav", "sha256": "be65576ab71c57e7752a0e8a06bfcf96836c13f84e7b79ae4007e5947c5e455e", "start_seconds_in_window": 451.74 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 454.02, "start": 453.74 }, "priority": "P0", "reference_segment_index": 98, "review_id": "timestamp-decisive-06", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_both_within_1s_outcome", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "对。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 291.93, "start": 291.69 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 292.049, "start": 291.849 }, "audio_clip": { "duration_seconds": 4.24, "end_seconds_in_window": 294.06, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-07-zhang-front.wav", "sha256": "88070736be6148233491f0a7cd24793ace939b315eee33694fe2d0d2a584a670", "start_seconds_in_window": 289.82 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 292.06, "start": 291.82 }, "priority": "P0", "reference_segment_index": 57, "review_id": "timestamp-decisive-07", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 293.45, "start": 293.13 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 293.569, "start": 293.249 }, "audio_clip": { "duration_seconds": 4.28, "end_seconds_in_window": 295.54, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-08-zhang-front.wav", "sha256": "39fcd6d4134c3aadd4682563106430ff1b7871c4e0b242eb30508e8279c216d0", "start_seconds_in_window": 291.26 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 293.54, "start": 293.26 }, "priority": "P0", "reference_segment_index": 59, "review_id": "timestamp-decisive-08", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "哦。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 330.37, "start": 330.13 }, "Volc_candidate": { "both_within_1s_of_machine_reference": false, "end": 330.45, "start": 327.89 }, "audio_clip": { "duration_seconds": 4.24, "end_seconds_in_window": 332.45, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-09-zhang-front.wav", "sha256": "9948a18b46981341eb7b9f14698bb39ae4118046464058715bfbae400579f582", "start_seconds_in_window": 328.21 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 330.45, "start": 330.21 }, "priority": "P0", "reference_segment_index": 66, "review_id": "timestamp-decisive-09", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 364.404, "start": 363.924 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 364.61, "start": 364.05 }, "audio_clip": { "duration_seconds": 4.56, "end_seconds_in_window": 366.61, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-10-zhang-front.wav", "sha256": "31b3a873a88603e7d2d7afe3ec1883938f7f6ef8046873c76661caaf703b1e90", "start_seconds_in_window": 362.05 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 364.61, "start": 364.05 }, "priority": "P0", "reference_segment_index": 79, "review_id": "timestamp-decisive-10", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯，对。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": null, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 397.86, "start": 397.62 }, "audio_clip": { "duration_seconds": 4.24, "end_seconds_in_window": 399.86, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-11-zhang-front.wav", "sha256": "c94ad08be22d2e1e8d556d062677ca4a36fbec54b4843147f314a565921e7450", "start_seconds_in_window": 395.62 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 397.86, "start": 397.62 }, "priority": "P0", "reference_segment_index": 87, "review_id": "timestamp-decisive-11", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": null, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 447.66, "start": 447.46 }, "audio_clip": { "duration_seconds": 4.2, "end_seconds_in_window": 449.66, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-12-zhang-front.wav", "sha256": "330d8ce3a02edc4d7cc1b51af9d714a80e2c1fbf98654c0d84d824007ba1bbd1", "start_seconds_in_window": 445.46 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 447.66, "start": 447.46 }, "priority": "P0", "reference_segment_index": 96, "review_id": "timestamp-decisive-12", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 467.172, "start": 466.932 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 467.22, "start": 466.98 }, "audio_clip": { "duration_seconds": 4.24, "end_seconds_in_window": 469.22, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-13-zhang-front.wav", "sha256": "2615b0d6185273501eb31b45caada3c575f57794707af4770688e1d0ad9a6e7f", "start_seconds_in_window": 464.98 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 467.22, "start": 466.98 }, "priority": "P0", "reference_segment_index": 101, "review_id": "timestamp-decisive-13", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 487.092, "start": 487.092 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 487.22, "start": 487.02 }, "audio_clip": { "duration_seconds": 4.2, "end_seconds_in_window": 489.22, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-14-zhang-front.wav", "sha256": "8356852a4d5aaffe8d5dfd48623f8be4a47424ec2e4cbe4e005fdd6d846bc025", "start_seconds_in_window": 485.02 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 487.22, "start": 487.02 }, "priority": "P0", "reference_segment_index": 106, "review_id": "timestamp-decisive-14", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": { "both_within_1s_of_machine_reference": true, "end": 504.649, "start": 504.489 }, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 504.78, "start": 504.54 }, "audio_clip": { "duration_seconds": 4.24, "end_seconds_in_window": 506.78, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-15-zhang-front.wav", "sha256": "23c68a83dbbf16cd96c1898d22a0838bedc2682450b6b41a5e79b41d4112a012", "start_seconds_in_window": 502.54 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 504.78, "start": 504.54 }, "priority": "P0", "reference_segment_index": 109, "review_id": "timestamp-decisive-15", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯。", "window_id": "front" }, { "MOSS_candidate": null, "Qwen_external_forced_aligner_candidate": null, "Volc_candidate": { "both_within_1s_of_machine_reference": true, "end": 511.1, "start": 510.98 }, "audio_clip": { "duration_seconds": 4.12, "end_seconds_in_window": 513.1, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/review-clips/timestamp-review-16-zhang-front.wav", "sha256": "77ccccfc6989c870c8703583de5404df5adf47e3fbcbf3cdd26fadedf9ee93ee", "start_seconds_in_window": 508.98 }, "decision_options": [ "machine_reference_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge", "Qwen_external_forced_aligner_boundary" ], "decision_required": "Listen for the lexical turn start and end. Choose a candidate or enter custom relative-to-episode start/end; do not use model identity as evidence.", "machine_reference_candidate": { "end": 511.1, "start": 510.98 }, "priority": "P0", "reference_segment_index": 111, "review_id": "timestamp-decisive-16", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 1, "selection_reason": "different_anchor_eligibility", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "嗯。", "window_id": "front" }, { "MOSS_candidate": { "end": 184.55, "start": 180.34 }, "Qwen_high_confidence_subset_alignment_candidate": { "both_within_1s_of_machine_reference": false, "end": 184.354, "start": 146.28 }, "Qwen_low_confidence_inclusive_alignment_candidate": { "both_within_1s_of_machine_reference": true, "end": 184.354, "start": 180.354 }, "Volc_candidate": { "end": 184.64, "start": 180.4 }, "alignment_boundary_delta_seconds": 34.074, "audio_clip": { "duration_seconds": 8.24, "end_seconds_in_window": 186.64, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v2/review-clips/timestamp-review-qwen-01-zhang-front.wav", "sha256": "f680f874d0678bb0b7cc13c91797049e4ff7b33859d155dbef5d74915e9e7765", "start_seconds_in_window": 178.4 }, "decision_options": [ "machine_reference_boundary", "Qwen_high_confidence_subset_alignment_boundary", "Qwen_low_confidence_inclusive_alignment_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge" ], "decision_required": "Listen for the lexical turn start and end. The two Qwen candidates come from the same raw output under different global text-alignment reference sets.", "machine_reference_candidate": { "end": 184.64, "start": 180.4 }, "priority": "P0", "reference_segment_index": 34, "review_id": "timestamp-qwen-alignment-sensitive-01", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 2, "selection_reason": "qwen_global_text_alignment_path_changes_boundary_hit", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "我看到这个文件，那一刻我觉得是一片空白的，就觉得说，是真的吗？", "window_id": "front" }, { "MOSS_candidate": { "end": 126.21, "start": 123.22 }, "Qwen_high_confidence_subset_alignment_candidate": { "both_within_1s_of_machine_reference": false, "end": 126.146, "start": 106.402 }, "Qwen_low_confidence_inclusive_alignment_candidate": { "both_within_1s_of_machine_reference": true, "end": 126.146, "start": 123.266 }, "Volc_candidate": { "end": 126.33, "start": 123.33 }, "alignment_boundary_delta_seconds": 16.864, "audio_clip": { "duration_seconds": 7.0, "end_seconds_in_window": 128.33, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v2/review-clips/timestamp-review-qwen-02-zhang-front.wav", "sha256": "ad16c6d4688e5bb0c7fcd9ebf6903e2201a071acb40d00e907955f7deb290c4d", "start_seconds_in_window": 121.33 }, "decision_options": [ "machine_reference_boundary", "Qwen_high_confidence_subset_alignment_boundary", "Qwen_low_confidence_inclusive_alignment_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge" ], "decision_required": "Listen for the lexical turn start and end. The two Qwen candidates come from the same raw output under different global text-alignment reference sets.", "machine_reference_candidate": { "end": 126.33, "start": 123.33 }, "priority": "P0", "reference_segment_index": 24, "review_id": "timestamp-qwen-alignment-sensitive-02", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 2, "selection_reason": "qwen_global_text_alignment_path_changes_boundary_hit", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "然后目前咱们的产品还没有开售，对吧？", "window_id": "front" }, { "MOSS_candidate": { "end": 349.34, "start": 348.93 }, "Qwen_high_confidence_subset_alignment_candidate": { "both_within_1s_of_machine_reference": false, "end": 356.763, "start": 356.683 }, "Qwen_low_confidence_inclusive_alignment_candidate": { "both_within_1s_of_machine_reference": true, "end": 349.163, "start": 349.003 }, "Volc_candidate": { "end": 349.29, "start": 349.09 }, "alignment_boundary_delta_seconds": 15.28, "audio_clip": { "duration_seconds": 4.2, "end_seconds_in_window": 351.29, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v2/review-clips/timestamp-review-qwen-03-zhang-front.wav", "sha256": "c3f5f818b2da80f02d93353fb7374f7752e2c490beeb12ea83ff7c89294b9115", "start_seconds_in_window": 347.09 }, "decision_options": [ "machine_reference_boundary", "Qwen_high_confidence_subset_alignment_boundary", "Qwen_low_confidence_inclusive_alignment_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge" ], "decision_required": "Listen for the lexical turn start and end. The two Qwen candidates come from the same raw output under different global text-alignment reference sets.", "machine_reference_candidate": { "end": 349.29, "start": 349.09 }, "priority": "P0", "reference_segment_index": 75, "review_id": "timestamp-qwen-alignment-sensitive-03", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 2, "selection_reason": "qwen_global_text_alignment_path_changes_boundary_hit", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "对。", "window_id": "front" }, { "MOSS_candidate": { "end": 348.82, "start": 344.93 }, "Qwen_high_confidence_subset_alignment_candidate": { "both_within_1s_of_machine_reference": false, "end": 356.283, "start": 344.923 }, "Qwen_low_confidence_inclusive_alignment_candidate": { "both_within_1s_of_machine_reference": true, "end": 348.603, "start": 344.923 }, "Volc_candidate": { "end": 348.85, "start": 345.01 }, "alignment_boundary_delta_seconds": 7.68, "audio_clip": { "duration_seconds": 7.84, "end_seconds_in_window": 350.85, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v2/review-clips/timestamp-review-qwen-04-zhang-front.wav", "sha256": "5cc9c592b85f4444464931c0b04e4e1c9db2634cc6625c2b372a8273a4f70b15", "start_seconds_in_window": 343.01 }, "decision_options": [ "machine_reference_boundary", "Qwen_high_confidence_subset_alignment_boundary", "Qwen_low_confidence_inclusive_alignment_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge" ], "decision_required": "Listen for the lexical turn start and end. The two Qwen candidates come from the same raw output under different global text-alignment reference sets.", "machine_reference_candidate": { "end": 348.85, "start": 345.01 }, "priority": "P0", "reference_segment_index": 74, "review_id": "timestamp-qwen-alignment-sensitive-04", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 2, "selection_reason": "qwen_global_text_alignment_path_changes_boundary_hit", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "但是如果你人生一直很顺利的时候，可能你就没时间，你也不会去思考这些事。", "window_id": "front" }, { "MOSS_candidate": { "end": 272.93, "start": 268.82 }, "Qwen_high_confidence_subset_alignment_candidate": { "both_within_1s_of_machine_reference": false, "end": 272.836, "start": 265.076 }, "Qwen_low_confidence_inclusive_alignment_candidate": { "both_within_1s_of_machine_reference": true, "end": 272.836, "start": 268.836 }, "Volc_candidate": { "end": 273.089, "start": 268.849 }, "alignment_boundary_delta_seconds": 3.76, "audio_clip": { "duration_seconds": 8.28, "end_seconds_in_window": 275.1, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v2/review-clips/timestamp-review-qwen-05-zhang-front.wav", "sha256": "9fa0365ca2b8f23466663889071a39bbbd4381481146c762cca7ffd9c8d8150c", "start_seconds_in_window": 266.82 }, "decision_options": [ "machine_reference_boundary", "Qwen_high_confidence_subset_alignment_boundary", "Qwen_low_confidence_inclusive_alignment_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge" ], "decision_required": "Listen for the lexical turn start and end. The two Qwen candidates come from the same raw output under different global text-alignment reference sets.", "machine_reference_candidate": { "end": 273.1, "start": 268.82 }, "priority": "P0", "reference_segment_index": 51, "review_id": "timestamp-qwen-alignment-sensitive-05", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 2, "selection_reason": "qwen_global_text_alignment_path_changes_boundary_hit", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "所以这个体量的减人已经不是这种正常的减法了，就是一个个会开下去。", "window_id": "front" }, { "MOSS_candidate": { "end": 368.25, "start": 364.62 }, "Qwen_high_confidence_subset_alignment_candidate": { "both_within_1s_of_machine_reference": false, "end": 368.084, "start": 362.644 }, "Qwen_low_confidence_inclusive_alignment_candidate": { "both_within_1s_of_machine_reference": true, "end": 368.084, "start": 364.644 }, "Volc_candidate": { "end": 368.33, "start": 364.77 }, "alignment_boundary_delta_seconds": 2.0, "audio_clip": { "duration_seconds": 7.56, "end_seconds_in_window": 370.33, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v2/review-clips/timestamp-review-qwen-06-zhang-front.wav", "sha256": "264c3eb4c8ecae571d97d589e946f283db385eb9d9a4caf4ceb7c2f4445b98f1", "start_seconds_in_window": 362.77 }, "decision_options": [ "machine_reference_boundary", "Qwen_high_confidence_subset_alignment_boundary", "Qwen_low_confidence_inclusive_alignment_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge" ], "decision_required": "Listen for the lexical turn start and end. The two Qwen candidates come from the same raw output under different global text-alignment reference sets.", "machine_reference_candidate": { "end": 368.33, "start": 364.77 }, "priority": "P0", "reference_segment_index": 80, "review_id": "timestamp-qwen-alignment-sensitive-06", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 2, "selection_reason": "qwen_global_text_alignment_path_changes_boundary_hit", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "所以那一年是什么样的思考，让你最后指向了要去做机器人？", "window_id": "front" }, { "MOSS_candidate": { "end": 481.94, "start": 477.07 }, "Qwen_high_confidence_subset_alignment_candidate": { "both_within_1s_of_machine_reference": false, "end": 481.892, "start": 476.132 }, "Qwen_low_confidence_inclusive_alignment_candidate": { "both_within_1s_of_machine_reference": true, "end": 481.892, "start": 477.092 }, "Volc_candidate": { "end": 482.06, "start": 477.14 }, "alignment_boundary_delta_seconds": 0.96, "audio_clip": { "duration_seconds": 8.92, "end_seconds_in_window": 484.06, "path": "runs/podcast-timestamp-provisional-agent-gold-20260721-v2/review-clips/timestamp-review-qwen-07-zhang-front.wav", "sha256": "295944f7ed76efc674f5a47282b4ea52a0714a2309472162ad115451cb81b4fe", "start_seconds_in_window": 475.14 }, "decision_options": [ "machine_reference_boundary", "Qwen_high_confidence_subset_alignment_boundary", "Qwen_low_confidence_inclusive_alignment_boundary", "MOSS_boundary", "Volc_boundary", "custom_start_end", "overlap_or_multiple_turns", "unable_to_judge" ], "decision_required": "Listen for the lexical turn start and end. The two Qwen candidates come from the same raw output under different global text-alignment reference sets.", "machine_reference_candidate": { "end": 482.06, "start": 477.14 }, "priority": "P0", "reference_segment_index": 104, "review_id": "timestamp-qwen-alignment-sensitive-07", "sample_id": "xiaoyuzhou_69dae02a_zhang_yi_home_robot", "schema_version": 2, "selection_reason": "qwen_global_text_alignment_path_changes_boundary_hit", "source_window_audio": { "path": "data/silver/podcast/embodied-intelligence-agent-consensus-v1/xiaoyuzhou_69dae02a_zhang_yi_home_robot/front/audio.wav", "sha256": "8bda8cb9caedc5614bb0c4bd01f41ad05b431088cf10101f29ec116b5d6a82af" }, "text": "有没有那么一个就关键时刻，你决定好了就是它了，我就是要 all in 开始干机器人了。", "window_id": "front" } ], "low_confidence_inclusive_ranking": [ "qwen3-asr-1.7b-plus-external-forced-aligner", "moss-transcribe-diarize-0.9b", "volcengine-current-asr" ], "new_qwen_review_item_count": 7, "new_review_clip_count": 7, "purpose": "three_system_rank_changing_timestamp_boundary_review", "ranking_changed_under_sensitivity": true, "reused_audio_only_no_new_clips": false, "reused_review_item_count": 16, "schema_version": 2, "source_queue": "runs/podcast-timestamp-provisional-agent-gold-20260721-v1/minimal-decisive-timestamp-review-queue.json", "source_queue_sha256": "b1586660eb9a49d381c9e86cbd8209438bd41dc38a445039ec608cf25219c6d9" }