[{"data":1,"prerenderedAt":271},["ShallowReactive",2],{"blog-post-podcast-audio-quality-study-2026":3},{"slug":4,"path":5,"title":6,"description":7,"publishedAt":8,"updatedAt":8,"author":9,"category":10,"tags":11,"featureImage":14,"ogImage":20,"canonical":21,"seoTitle":22,"seoDescription":22,"keywords":23,"readingTimeMinutes":28,"wordCount":29,"bodySegments":30,"hasMath":111,"headings":112,"mediaClips":137,"showFeatureImage":111,"leadFigureHtml":22,"relatedPosts":146,"citations":270},"podcast-audio-quality-study-2026","/blog/podcast-audio-quality-study-2026/","We measured the sound of 1,438 podcasts: the charts don't sound better than the long tail","A measured study of speech quality in 3,525 podcast episodes: chart shows match long-tail shows, genres differ, and the same shows have barely changed since 2017.","2026-11-06T14:51:00Z","Nathan Harmon","restoration",[12,13],"Podcasting","Research",{"src":15,"alt":16,"widthPixels":17,"heightPixels":18,"srcset":19},"/media/blog/podcast-audio-quality-study-2026/podcast-quality-by-genre.webp","Bar chart of median speech quality score by podcast genre, from Technology at 4.27 down to Kids and Family at 3.54",1430,1040,"/media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-genre-480w.webp 480w, /media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-genre-800w.webp 800w, /media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-genre-1200w.webp 1200w, /media/blog/podcast-audio-quality-study-2026/podcast-quality-by-genre.webp 1430w",{"src":15,"alt":16,"widthPixels":17,"heightPixels":18,"srcset":19},"https://diffio.ai/blog/podcast-audio-quality-study-2026/",null,[24,25,26,27],"podcast statistics 2026","podcast audio quality","podcast audio quality statistics","podcast sound quality by genre",6,1409,[31,108],{"kind":32,"comparison":33},"audio-comparison",{"passageName":34,"originalSourceNote":35,"originalSrc":36,"restoredSrc":37,"originalWaveformPeaks":38,"restoredWaveformPeaks":84},"Rev. W. J. Faulkner, score 4.40","AFS 6361A, Nashville, December 1941","/media/blog/pearl-harbor-1941-man-on-the-street-restored/nashville-original.mp3","/media/blog/pearl-harbor-1941-man-on-the-street-restored/nashville-restored.mp3",[39,40,41,42,43,44,45,39,46,47,47,46,46,47,48,49,50,51,52,46,53,54,55,56,41,57,39,58,59,55,60,46,61,49,62,43,46,63,61,64,39,65,66,67,53,46,68,69,55,62,70,39,71,72,73,46,46,74,42,53,55,60,51,46,75,76,70,46,49,63,46,77,78,72,79,46,80,78,65,46,46,81,82,46,78,77,68,46,50,50,57,43,83,51,67,53],0.07,0.49,0.71,0.28,0.41,0.17,0.36,0.08,0.72,0.75,0.35,0.52,0.18,0.5,0.09,0.89,0.59,0.62,0.42,0.06,0.95,0.23,0.83,0.51,0.14,0.56,0.82,0.48,0.84,0.45,0.22,0.29,0.65,0.44,0.25,0.91,0.68,0.57,0.77,1,0.37,0.81,0.47,0.26,0.21,[58,72,85,86,87,88,49,58,58,71,89,58,58,71,47,90,66,91,92,58,58,93,50,64,89,94,58,58,54,50,69,58,77,95,92,96,58,97,47,98,58,48,87,99,58,58,87,91,52,98,100,58,76,79,101,58,58,65,60,58,102,83,97,58,55,66,73,58,42,46,58,56,78,57,95,58,103,78,104,58,58,72,60,58,78,105,92,58,102,98,92,106,60,107,65,58],0.61,0.3,0.4,0.13,0.64,0.32,0.15,0.43,0.78,0.38,0.31,0.33,0.11,0.46,0.67,0.24,0.2,0.55,0.76,0.79,0.69,0.39,0.16,{"kind":109,"html":110},"html","\u003Cp>The two players above are what this study's scale sounds like: a \u003Ca href=\"/blog/pearl-harbor-1941-man-on-the-street-restored/\">December 1941 Library of Congress disc\u003C/a> scores 2.50 as it survives and 4.40 after cleanup with Diffio. The podcasts below were measured as published, with no cleanup of any kind.\u003C/p>\n\u003Cp>Podcasters talk about audio quality constantly, and most of the numbers quoted about it are survey answers: what listeners say puts them off. I couldn't find anyone publishing what podcasts actually sound like, measured, across a large sample. So at Diffio AI we measured it.\u003C/p>\n\u003Cp>We ran a speech quality model over 3,525 English-language episodes from 1,438 shows: shows on Apple's US genre charts, shows outside them, and a panel of chart shows traced back to 2017.\u003C/p>\n\u003Ch2 id=\"the-numbers\">The numbers\u003C/h2>\n\u003Cdiv class=\"blog-table blog-table--notes\" role=\"region\" aria-label=\"Table 1: Finding, Figures\" tabindex=\"0\">\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>Finding\u003C/th>\n\u003Cth>Figures\u003C/th>\n\u003C/tr>\n\u003C/thead>\n\u003Ctbody>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>The charts and the long tail sound the same.\u003C/strong>\u003C/td>\n\u003Ctd>The latest episode of a chart show has a median score of 4.06 out of 5. An active show outside the charts also has a median of 4.06.\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>About 1 in 5 chart episodes score below 3.5\u003C/strong>\u003C/td>\n\u003Ctd>19.2% below 3.5, and about 1 in 14 below 3.0 (7.0%). For long-tail shows: 15.8% and 4.4%.\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Chart position barely matters.\u003C/strong>\u003C/td>\n\u003Ctd>Top 10 of their genre: median 4.09. Ranked 51 to 100: 4.05. The top 10 do have fewer weak episodes: 12.7% below 3.5, against 20.1%.\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Genre matters more.\u003C/strong>\u003C/td>\n\u003Ctd>Technology has the highest median at 4.27. Kids & Family has the lowest at 3.54.\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Ten years have changed little.\u003C/strong>\u003C/td>\n\u003Ctd>The same shows moved from 3.85 in 2017 to 4.00 in 2026, on a five-point scale.\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>The pandemic left no mark.\u003C/strong>\u003C/td>\n\u003Ctd>From 2019 to 2020, when many shows moved to remote recording, the same shows changed by +0.05 (95% interval -0.05 to +0.14), which can't be told from no change.\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>\u003Cstrong>Bitrate tells you nothing about the score.\u003C/strong>\u003C/td>\n\u003Ctd>Episodes at 64 kbps or less: median 4.06. Episodes above 200 kbps: 4.07.\u003C/td>\n\u003C/tr>\n\u003C/tbody>\n\u003C/table>\n\u003C/div>\u003Ch2 id=\"what-the-score-means\">What the score means\u003C/h2>\n\u003Cp>The score comes from SCOREQ, a no-reference speech quality model: it listens to speech and predicts a quality score from 1 to 5 without needing a clean copy to compare against. It's a model's prediction and not a panel of human listeners, so read it as a consistent yardstick.\u003C/p>\n\u003Cp>For scale: the 59 real-world noisy recordings in \u003Ca href=\"/speech-enhancer-benchmark/\">our speech enhancer benchmark\u003C/a>, full of echo, wind, crowds and hum, average 2.31 before any cleanup. A typical podcast at about 4 is well clear of that. A score under 3 is closer to that territory than to a good studio recording. The disc at the top of this page gives you both ends by ear.\u003C/p>\n\u003Ch2 id=\"genre-talk-shows-score-highest-produced-shows-lowest\">Genre: talk shows score highest, produced shows lowest\u003C/h2>\n\u003Cfigure class=\"blog-figure blog-figure--image\">\u003Cdiv class=\"blog-figure__frame\">\u003Cimg src=\"/media/blog/podcast-audio-quality-study-2026/podcast-quality-by-genre.webp\" srcset=\"/media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-genre-480w.webp 480w, /media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-genre-800w.webp 800w, /media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-genre-1200w.webp 1200w, /media/blog/podcast-audio-quality-study-2026/podcast-quality-by-genre.webp 1430w\" sizes=\"(max-width: 760px) calc(100vw - 40px), 720px\" alt=\"Bar chart of median speech quality score by podcast genre, from Technology at 4.27 down to Kids and Family at 3.54\" width=\"1430\" height=\"1040\" loading=\"lazy\" decoding=\"async\">\u003C/div>\u003Cfigcaption>Median score of the latest episode by Apple chart genre. Lines show the 95% interval.\u003C/figcaption>\u003Ca class=\"blog-figure__full\" href=\"/media/blog/podcast-audio-quality-study-2026/podcast-quality-by-genre.webp\">View full size\u003Cspan class=\"blog-visually-hidden\">: Bar chart of median speech quality score by podcast genre, from Technology at 4.27 down to Kids and Family at 3.54\u003C/span>\u003C/a>\u003C/figure>\n\u003Cdiv class=\"blog-table blog-table--wide blog-table--fit\" role=\"region\" aria-label=\"Table 2: Genre, Episodes, Median, Below 3.5\" tabindex=\"0\">\n\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>Genre\u003C/th>\n\u003Cth style=\"text-align:right\">Episodes\u003C/th>\n\u003Cth style=\"text-align:right\">Median\u003C/th>\n\u003Cth style=\"text-align:right\">Below 3.5\u003C/th>\n\u003C/tr>\n\u003C/thead>\n\u003Ctbody>\n\u003Ctr>\n\u003Ctd>Technology\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">51\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.27\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">5.9%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Health & Fitness\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">39\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.24\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">7.7%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>News\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">50\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.23\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">8.0%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Business\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">52\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.21\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">5.8%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Religion & Spirituality\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">37\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.17\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">13.5%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Science\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">62\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.15\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">17.7%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Education\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">50\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.15\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">12.0%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Government\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">46\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.10\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">13.0%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Arts\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">40\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.08\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">12.5%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>History\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">51\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.07\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">29.4%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Leisure\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">51\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">4.06\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">25.5%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Society & Culture\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">44\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">3.99\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">20.5%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Sports\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">44\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">3.89\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">20.5%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Comedy\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">55\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">3.89\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">23.6%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>TV & Film\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">60\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">3.84\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">23.3%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>True Crime\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">50\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">3.84\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">14.0%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Fiction\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">49\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">3.68\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">40.8%\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Kids & Family\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">49\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">3.54\u003C/td>\n\u003Ctd style=\"text-align:right\" class=\"blog-table__number\">46.9%\u003C/td>\n\u003C/tr>\n\u003C/tbody>\n\u003C/table>\n\u003C/div>\u003Cp>These genres aren't equally comparable. The model scores speech, so music and sound effects under a voice can lower the score even when the production is excellent. We didn't separate those cases, and that may explain much of the low scores for Fiction and Kids & Family. Treat a low score in those genres as "the voice has company", not as "badly made". Technology, News and Business, on the other hand, are mostly plain talk, so their scores say more about the recording itself.\u003C/p>\n\u003Ch2 id=\"the-same-shows-2017-to-2026\">The same shows, 2017 to 2026\u003C/h2>\n\u003Cfigure class=\"blog-figure blog-figure--image\">\u003Cdiv class=\"blog-figure__frame\">\u003Cimg src=\"/media/blog/podcast-audio-quality-study-2026/podcast-quality-by-year.webp\" srcset=\"/media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-year-480w.webp 480w, /media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-year-800w.webp 800w, /media/blog/podcast-audio-quality-study-2026/responsive/podcast-quality-by-year-1200w.webp 1200w, /media/blog/podcast-audio-quality-study-2026/podcast-quality-by-year.webp 1430w\" sizes=\"(max-width: 760px) calc(100vw - 40px), 720px\" alt=\"Line chart of mean speech quality score for the same podcasts from 2017 to 2026, nearly flat, rising from 3.85 to 4.00\" width=\"1430\" height=\"728\" loading=\"lazy\" decoding=\"async\">\u003C/div>\u003Cfigcaption>476 chart shows with back catalogs, one episode a year, adjusted for which shows each year includes.\u003C/figcaption>\u003Ca class=\"blog-figure__full\" href=\"/media/blog/podcast-audio-quality-study-2026/podcast-quality-by-year.webp\">View full size\u003Cspan class=\"blog-visually-hidden\">: Line chart of mean speech quality score for the same podcasts from 2017 to 2026, nearly flat, rising from 3.85 to 4.00\u003C/span>\u003C/a>\u003C/figure>\n\n\u003Cp>Microphones, remote recording platforms and AI cleanup tools all changed a lot over this period, so I expected to see it in the numbers. The line is almost flat, rising about 0.012 points a year. Comparing the same shows, 2019 to 2025 is +0.03 (95% interval -0.07 to +0.12), and 2019 to 2026 is +0.14 (+0.04 to +0.23). Most of the rise is in the last year, and I'd wait for another year of data before calling that a trend, partly because 2026 is the one year in which we took each show's newest episode.\u003C/p>\n\u003Cp>A flat line doesn't mean nothing improved. It could mean the shows that were already on the charts in 2017 had already solved this, which is a limit of tracing chart shows backward.\u003C/p>\n\u003Ch2 id=\"how-we-measured-it\">How we measured it\u003C/h2>\n\u003Cdiv class=\"blog-table blog-table--compact\" role=\"region\" aria-label=\"Table 3: Parameter, Value\" tabindex=\"0\">\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>Parameter\u003C/th>\n\u003Cth>Value\u003C/th>\n\u003C/tr>\n\u003C/thead>\n\u003Ctbody>\n\u003Ctr>\n\u003Ctd>Shows\u003C/td>\n\u003Ctd>1,438: Apple's US top 100 in each of 18 genres, plus active shows outside them\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Episodes\u003C/td>\n\u003Ctd>3,525 scored, of 3,669 drawn from 6,650 planned\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Windows\u003C/td>\n\u003Ctd>10,418: three 24-second windows per episode\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Intervals\u003C/td>\n\u003Ctd>95% bootstrap\u003C/td>\n\u003C/tr>\n\u003Ctr>\n\u003Ctd>Measured\u003C/td>\n\u003Ctd>November 6, 2026\u003C/td>\n\u003C/tr>\n\u003C/tbody>\n\u003C/table>\n\u003C/div>\u003Cdetails class=\"blog-details\">\u003Csummary>How the shows, episodes and windows were chosen\u003C/summary>\u003Cdiv class=\"blog-details__body\">\u003Cul>\n\u003Cli>\u003Cstrong>Shows.\u003C/strong> Apple's US top 100 in each of 18 genres, plus shows found through Apple's podcast search on 93 everyday words that aren't on those charts and released an episode in the last 120 days. We left out the Music genre, video feeds and feeds with a non-English language tag. The long tail here means active shows that search can find, not a random sample of every podcast.\u003C/li>\n\u003Cli>\u003Cstrong>Episodes.\u003C/strong> The newest full episode of each show (8 minutes to 4 hours). For a random panel of chart shows, also the episode nearest September 30 of each year from 2017 to 2025. We planned 6,650 episodes in random order and stopped at 3,669, so the sample is a random part of the plan; 3,525 had enough speech to score.\u003C/li>\n\u003Cli>\u003Cstrong>Measurement.\u003C/strong> Three 24-second windows per episode, at 30%, 50% and 70% of the running time, 10,418 windows in all, level-matched to the same loudness, with windows under 70% speech dropped. An episode's score is the median of its windows.\u003C/li>\n\u003Cli>\u003Cstrong>Intervals.\u003C/strong> 95% bootstrap intervals. The year figures are adjusted for each show's own average, so a change in which shows are in the sample doesn't look like a change over time.\u003C/li>\n\u003C/ul>\u003C/div>\u003C/details>\n\u003Ch2 id=\"what-this-cant-tell-you\">What this can't tell you\u003C/h2>\n\u003Cul>\n\u003Cli>\u003Cstrong>It doesn't rank shows, so we name none.\u003C/strong> Three windows of one episode are a noisy read on a show: the first and last windows in an episode correlate at only 0.49, and a show's score in one year correlates with the next at 0.44. Averages over hundreds of episodes hold up. A single show's number doesn't.\u003C/li>\n\u003Cli>\u003Cstrong>Ads get in.\u003C/strong> A window can land on a host-read or dynamically inserted ad, and old episodes can carry new ads. Taking the median of three windows reduces this but doesn't remove it, and it may flatten differences between years.\u003C/li>\n\u003Cli>\u003Cstrong>It isn't about taste.\u003C/strong> A low score doesn't mean a show is bad or that listeners mind; listener behavior wasn't measured.\u003C/li>\n\u003C/ul>\n\u003Ch2 id=\"the-data\">The data\u003C/h2>\n\u003Cp>Both result tables can be downloaded here. Please cite Diffio AI if you use them: \u003Ca href=\"/media/blog/podcast-audio-quality-study-2026/podcast-quality-by-genre.csv\">genre results (CSV)\u003C/a> and \u003Ca href=\"/media/blog/podcast-audio-quality-study-2026/podcast-quality-by-year.csv\">year results (CSV)\u003C/a>. We kept scores, not audio: every window was deleted after it was measured.\u003C/p>\n\u003Ch2 id=\"our-stake-in-this\">Our stake in this\u003C/h2>\n\u003Cp>Diffio makes speech cleanup software, so we have an interest in audio quality being a topic. No Diffio tool touched the podcast audio we measured, and nothing here tests or compares cleanup tools; the only Diffio output on this page is the 1941 example at the top. Our comparison of those is \u003Ca href=\"/speech-enhancer-benchmark/\">the speech enhancer benchmark\u003C/a>, and it shows where we lost. If you want a number I didn't publish or want to check a figure, write to me at nathan@trydiffio.com.\u003C/p>",false,[113,116,119,122,125,128,131,134],{"id":114,"text":115,"label":115},"the-numbers","The numbers",{"id":117,"text":118,"label":118},"what-the-score-means","What the score means",{"id":120,"text":121,"label":121},"genre-talk-shows-score-highest-produced-shows-lowest","Genre: talk shows score highest, produced shows lowest",{"id":123,"text":124,"label":124},"the-same-shows-2017-to-2026","The same shows, 2017 to 2026",{"id":126,"text":127,"label":127},"how-we-measured-it","How we measured it",{"id":129,"text":130,"label":130},"what-this-cant-tell-you","What this can't tell you",{"id":132,"text":133,"label":133},"the-data","The data",{"id":135,"text":136,"label":136},"our-stake-in-this","Our stake in this",[138,143],{"kind":139,"src":36,"name":140,"caption":22,"posterSrc":22,"comparisonRole":141,"durationSeconds":142},"audio","Original (AFS 6361A, Nashville, December 1941), Rev. W. J. Faulkner, score 2.50","original",77.5,{"kind":139,"src":37,"name":144,"caption":22,"posterSrc":22,"comparisonRole":145,"durationSeconds":142},"Restored with Diffio, Rev. W. J. Faulkner, score 4.40","restored",[147,181,212,240],{"slug":148,"path":149,"title":150,"description":151,"publishedAt":152,"updatedAt":153,"author":9,"category":10,"tags":154,"featureImage":157,"ogImage":163,"canonical":168,"seoTitle":169,"seoDescription":170,"keywords":171,"readingTimeMinutes":179,"wordCount":180,"relatedNote":22},"jfk-fort-worth-1963-last-speeches-restored","/blog/jfk-fort-worth-1963-last-speeches-restored/","Hear JFK's last two speeches, Fort Worth, November 22, 1963, restored","President Kennedy's rally in the rain and his Chamber of Commerce breakfast on the morning of November 22, 1963, from the JFK Library's recordings, before and after restoration with Diffio AI.","2026-11-05T01:30:00Z","2026-11-05T07:21:06Z",[155,156],"Audio restoration","History",{"src":158,"alt":159,"widthPixels":160,"heightPixels":161,"srcset":162},"/media/blog/jfk-fort-worth-1963-last-speeches-restored/jfk-rally-spectrogram-v2.webp","Spectrograms of President Kennedy's Fort Worth rally recording, November 22, 1963, original and restored with Diffio AI",940,616,"/media/blog/jfk-fort-worth-1963-last-speeches-restored/responsive/jfk-rally-spectrogram-v2-480w.webp 480w, /media/blog/jfk-fort-worth-1963-last-speeches-restored/responsive/jfk-rally-spectrogram-v2-800w.webp 800w, /media/blog/jfk-fort-worth-1963-last-speeches-restored/jfk-rally-spectrogram-v2.webp 940w",{"src":164,"alt":150,"widthPixels":165,"heightPixels":166,"srcset":167},"/media/blog/jfk-fort-worth-1963-last-speeches-restored/social-card.png",1200,630,"/media/blog/jfk-fort-worth-1963-last-speeches-restored/responsive/social-card-480w.webp 480w, /media/blog/jfk-fort-worth-1963-last-speeches-restored/responsive/social-card-800w.webp 800w, /media/blog/jfk-fort-worth-1963-last-speeches-restored/social-card.png 1200w","https://diffio.ai/blog/jfk-fort-worth-1963-last-speeches-restored/","JFK's last speeches restored: Fort Worth, November 22, 1963","Hear President Kennedy's Fort Worth rally and breakfast speech from November 22, 1963, original and restored side by side, from JFK Library recordings.",[172,173,174,175,176,177,178],"JFK last speech","Kennedy Fort Worth speech","November 22, 1963","Hotel Texas rally","Fort Worth Chamber of Commerce breakfast","JFK Library White House audio","historical audio restoration",3,798,{"slug":182,"path":183,"title":184,"description":185,"publishedAt":186,"updatedAt":187,"author":9,"category":10,"tags":188,"featureImage":189,"ogImage":195,"canonical":200,"seoTitle":201,"seoDescription":202,"keywords":203,"readingTimeMinutes":210,"wordCount":211,"relatedNote":22},"pearl-harbor-1941-man-on-the-street-restored","/blog/pearl-harbor-1941-man-on-the-street-restored/","The day after Pearl Harbor, in Americans' own voices: 1941 street interviews, restored","On December 8, 1941, the Library of Congress asked fieldworkers across the US to record what ordinary people thought of the war. Twelve excerpts from those discs, original and restored.","2026-11-04T22:13:33Z","2026-11-05T06:54:57Z",[155,156],{"src":190,"alt":191,"widthPixels":192,"heightPixels":193,"srcset":194},"/media/blog/pearl-harbor-1941-man-on-the-street-restored/card.webp","Spectrogram of Rev. W. J. Faulkner's December 1941 recording in Nashville, original above and restored below with Diffio AI",1300,696,"/media/blog/pearl-harbor-1941-man-on-the-street-restored/responsive/card-480w.webp 480w, /media/blog/pearl-harbor-1941-man-on-the-street-restored/responsive/card-800w.webp 800w, /media/blog/pearl-harbor-1941-man-on-the-street-restored/card.webp 1300w",{"src":196,"alt":191,"widthPixels":197,"heightPixels":198,"srcset":199},"/media/blog/pearl-harbor-1941-man-on-the-street-restored/feature.webp",1400,800,"/media/blog/pearl-harbor-1941-man-on-the-street-restored/responsive/feature-480w.webp 480w, /media/blog/pearl-harbor-1941-man-on-the-street-restored/responsive/feature-800w.webp 800w, /media/blog/pearl-harbor-1941-man-on-the-street-restored/responsive/feature-1200w.webp 1200w, /media/blog/pearl-harbor-1941-man-on-the-street-restored/feature.webp 1400w","https://diffio.ai/blog/pearl-harbor-1941-man-on-the-street-restored/","Pearl Harbor 1941 man-on-the-street interviews, restored","Hear Americans react to Pearl Harbor in December 1941: Library of Congress street interviews from across the US, original and restored side by side.",[204,205,206,207,208,209,178],"Pearl Harbor man-on-the-street interviews","After the Day of Infamy","December 1941 oral history","Alan Lomax field recordings","American Folklife Center","Dear Mr. President 1942",8,1775,{"slug":213,"path":214,"title":215,"description":216,"publishedAt":217,"updatedAt":218,"author":9,"category":10,"tags":219,"featureImage":220,"ogImage":226,"canonical":227,"seoTitle":228,"seoDescription":229,"keywords":230,"readingTimeMinutes":238,"wordCount":239,"relatedNote":22},"pershing-1918-battlefields-of-france-restored","/blog/pershing-1918-battlefields-of-france-restored/","Hear General Pershing's 1918 message from the battlefields of France, restored","A 33-second disc General John J. Pershing recorded in France in March 1918, restored for Veterans Day, with the original alongside.","2026-11-02T08:40:45Z","2026-11-05T23:08:28Z",[155,156],{"src":221,"alt":222,"widthPixels":223,"heightPixels":224,"srcset":225},"/media/blog/pershing-1918-battlefields-of-france-restored/feature.webp","Waveform of Gen. John J. Pershing's 1918 recording \"From the Battlefields of France,\" restored with Diffio AI",1280,720,"/media/blog/pershing-1918-battlefields-of-france-restored/responsive/feature-480w.webp 480w, /media/blog/pershing-1918-battlefields-of-france-restored/responsive/feature-800w.webp 800w, /media/blog/pershing-1918-battlefields-of-france-restored/feature.webp 1280w",{"src":221,"alt":222,"widthPixels":223,"heightPixels":224,"srcset":225},"https://diffio.ai/blog/pershing-1918-battlefields-of-france-restored/","General Pershing's 1918 message from France, restored","General Pershing's 1918 Nation's Forum disc, recorded in France while he led the American Expeditionary Forces, restored beside the original.",[231,232,233,234,235,236,237],"John J. Pershing recording","From the Battlefields of France","Nation's Forum records","World War I audio","American Leaders Speak","Veterans Day","acoustic recording restoration",4,827,{"slug":241,"path":242,"title":243,"description":244,"publishedAt":245,"updatedAt":187,"author":9,"category":246,"tags":247,"featureImage":249,"ogImage":255,"canonical":258,"seoTitle":259,"seoDescription":260,"keywords":261,"readingTimeMinutes":238,"wordCount":269,"relatedNote":22},"save-a-loved-ones-voice-recording","/blog/save-a-loved-ones-voice-recording/","How to save a loved one's voice from a voicemail or old family video, and clean it up","A practical guide to saving a loved one's voice from a phone voicemail, a cassette or an old family video, keeping the original safe, and making the voice easier to hear.","2026-11-04T09:00:00Z","guides",[155,248],"Family history",{"src":250,"alt":251,"widthPixels":252,"heightPixels":253,"srcset":254},"/media/blog/save-a-loved-ones-voice-recording/lead-diagram.webp","Diagram of four steps: save the recording to a file, keep the original in two places, clean up a copy and compare it by ear, share the clean copy with family.",1440,810,"/media/blog/save-a-loved-ones-voice-recording/responsive/lead-diagram-480w.webp 480w, /media/blog/save-a-loved-ones-voice-recording/responsive/lead-diagram-800w.webp 800w, /media/blog/save-a-loved-ones-voice-recording/responsive/lead-diagram-1200w.webp 1200w, /media/blog/save-a-loved-ones-voice-recording/lead-diagram.webp 1440w",{"src":256,"alt":243,"widthPixels":165,"heightPixels":166,"srcset":257},"/media/blog/save-a-loved-ones-voice-recording/social-card.png","/media/blog/save-a-loved-ones-voice-recording/responsive/social-card-480w.webp 480w, /media/blog/save-a-loved-ones-voice-recording/responsive/social-card-800w.webp 800w, /media/blog/save-a-loved-ones-voice-recording/social-card.png 1200w","https://diffio.ai/blog/save-a-loved-ones-voice-recording/","How to save a loved one's voice from a voicemail or video","How to save a loved one's voice from a voicemail, cassette or old family video, keep the original safe, and make the voice easier to hear.",[262,263,264,265,266,267,268],"save a voicemail","save voicemail from iPhone","export voicemail on Android","digitize cassette tapes","transfer VHS family videos","back up family recordings","clean up old voice recordings",1028,[],1794173866912]