CentificAIResearch/audio-loop

Audio Loop: Long-Audio Reasoning Benchmarks

An evaluation set for long-form audio reasoning. It accompanies the paper AudioLoop: Program-Guided Iterative Reasoning Over Long-Form Speech (NeurIPS 2026 Workshop on Long Context Foundation Models), where it is referred to as IQ2-QA.

Config	Task	Audio source	Items
iq2_qa	Multi-hop QA over debates	Recorded live debates	13

Load annotations

from datasets import load_dataset

iq2 = load_dataset("CentificAIResearch/audio-loop", "iq2_qa", split="test")

Each row records the audio file used in the paper (file) and the podcast episode it came from (source).

Obtaining the audio

The debate audio is not redistributed here: it is copyrighted by Open to Debate Foundation. Each row's source field identifies the exact podcast episode used, so the audio can be obtained from the publisher:

row = iq2[0]
print(row["source"]["episode_title"], row["source"]["episode_guid"])
print(row["source"]["episode_audio_url"])   # audio URL as listed in the feed

Match episodes by episode_guid in the public RSS feed (feed_url). Podcast hosts insert advertisements dynamically, so a fresh download can differ from the paper's file by roughly a minute in length.

---

iq2_qa (IQ2-QA)

Overview

Multi-hop reasoning queries over 13 complete Oxford-style debate recordings (~50–60 min each). In each debate, one or two debaters per side argue FOR and AGAINST a motion, in a live-audience format.

Source

The Open to Debate podcast (formerly Intelligence Squared U.S.): https://podcasts.apple.com/us/podcast/open-to-debate/id216713308. Audio was downloaded from the public RSS feed in April 2026. Questions and reference answers were drafted with Gemini 3 Pro; reference answers are framed in terms of the FOR and AGAINST sides of each motion.

License note

Annotations (queries, reference answers, metadata) are released under CC-BY-4.0. The debate audio is copyrighted by Open to Debate Foundation and is not included; obtain it from the publisher's feed under its terms.

Schema (metadata.jsonl, 1 row per debate)

{
  "debate_id": "iq2_161",
  "file": "161_Should_the_FTC_Ban_Non-compete_Clauses_.mp3",
  "source": {
    "podcast": "Open to Debate (formerly Intelligence Squared U.S.)",
    "feed_url": "https://feeds.megaphone.fm/PNP1207584390",
    "episode_title": "Should the FTC Ban Non-compete Clauses?",
    "episode_guid": "7a72b250-c9ac-11ed-9d3f-bf45543f63e9",
    "episode_pub_date": "2023-03-24",
    "episode_audio_url": "https://www.podtrac.com/pts/redirect.mp3/traffic.megaphone.fm/IS6200446555.mp3?updated=1679598682"
  },
  "album": "Intelligence Squared U.S. Debates",
  "is_iq2": true,
  "is_debate": true,
  "duration_minutes": 51.3,
  "debate_topic": "Should the FTC Ban Non-compete Clauses?",
  "queries": [
    {
      "query_id": "iq2_161_A",
      "query_type": "claim_rebuttal_counter",
      "query": "What was the main argument made by the FOR side ...",
      "hop_count": 3,
      "requires_acoustic_cues": false,
      "difficulty": "medium",
      "reference_answer": "..."
    }
  ]
}

Query types

Type	Count	Description
claim_rebuttal_counter	13	FOR claim → AGAINST rebuttal → FOR counter (3-hop)
theme_comparison	13	Compare both sides on a shared theme (2-hop)
tone_disagreement	13	Identify the most heated exchange (2-hop, acoustic cues)

39 queries total across 13 debates (3 per debate).

Format

• Audio: not included (MP3 episodes from the publisher's feed)
• Annotations: JSON Lines, one row per debate

Stats

Stat	Value
Debates	13
Queries	39
Per-debate duration	~50 – 60 min
Total audio runtime	~11 hours

---

Licensing summary

Item	License
iq2_qa/ annotations	CC-BY-4.0
iq2_qa audio (not included)	Copyright Open to Debate Foundation; obtain from the publisher

---

Changes

• October 2026: The aae_tts config (AAE-TTS, a two-speaker speech rendering of the Argument Annotated Essays corpus, version 2) was removed. The AAE2 license does not permit redistribution of the corpus, so AAE-TTS is used for evaluation in the paper but is not released.
• October 2026: The IQ2-QA debate audio was removed, because it is copyrighted by Open to Debate Foundation. Each row now records the source episode (source), so the exact files used in the paper can be traced.