30-minute podcast: transcript plus summary voiceover cost by vendor
Transcribing 30 minutes and reading a 4,500-character summary aloud costs $0.12 on xAI, $0.20 on OpenAI, $0.21 on ElevenLabs and $0.51 on Sume. Math inside.
Transcribing a 30-minute podcast and reading a 4,500-character summary back as audio costs about $0.12 on xAI, $0.20 on OpenAI, $0.21 on ElevenLabs and $0.51 on Sume. The transcript is the smaller line on every vendor except Sume, where speech-to-text is $0.01 per audio minute and a 10-minute job cap means three jobs.
Sources are the xAI, OpenAI and ElevenLabs pricing pages (read 2026-10-04) and Sume's API pricing.
| Vendor and models | Transcript | Voiceover | Total |
|---|---|---|---|
| xAI STT $0.10/h + TTS $15/M | $0.05 | $0.0675 | $0.1175 |
| OpenAI gpt-transcribe $0.0045/min + tts-1 $15/M | $0.135 | $0.0675 | $0.2025 |
| ElevenLabs Scribe v2 $0.22/h + v4 sale $0.022/1K | $0.11 | $0.099 | $0.209 |
| Sume STT $0.01/min + TTS $0.0475/1K | $0.30 | $0.21375 | $0.51375 |
The line items
Transcript: xAI REST is $0.10 per hour, so half an hour is $0.05. OpenAI gpt-transcribe is $0.0045 per minute, 30 x $0.0045 = $0.135. ElevenLabs Scribe v2 is $0.22 per hour, so $0.11. Sume is $0.01 per audio minute, 30 x $0.01 = $0.30.
Voiceover: xAI and OpenAI tts-1 both list $15 per 1M characters, 4,500 x $0.000015 = $0.0675. ElevenLabs v4 is $0.022 per 1K on sale, 4.5 x $0.022 = $0.099, and $0.36 (4.5 x $0.08) at list after Oct 12. Sume is 4.5 x $0.0475 = $0.21375.
Why Sume needs three transcription jobs
A Sume speech-to-text job is capped at 10 minutes of audio, so a 30-minute file is split into three 10-minute jobs at $0.10 each. The total is the same $0.30 as the per-minute math, and a failed slice is refunded without touching the other two.
Cut at silence or a topic break and keep the offsets, so you can stitch the three transcripts back in order.
- Slice 1: 0:00-10:00, $0.10
- Slice 2: 10:00-20:00, $0.10
- Slice 3: 20:00-30:00, $0.10
Steps
- Transcribe the episode and review the text.
- Write or generate a summary of roughly 4,500 characters (about 700 words).
- Send the summary to TTS and attach the audio to your show notes or social cut.
- If you also need a video teaser, a Timeline render of up to one minute is $0.10 with the narration as the spine.
Ten episodes
Ten episodes of this size cost $1.175 on xAI, $2.025 on OpenAI, $2.09 on ElevenLabs and $5.1375 on Sume at the rates above. The spread is $3.96 across ten episodes, so the extra work of running a second vendor is hard to justify on price alone.
Caveats
Accuracy, speaker labels, languages and timestamps differ between transcription models, and the cheapest row is not automatically the best fit for noisy audio. All four totals stay under a dollar, so quality and workflow usually decide. If your show is weekly, the yearly cost of 52 episodes at the lowest and highest rows is 52 x $0.1175 = $6.11 and 52 x $0.51375 = $26.72.
Sources
More in Comparisons
- One podcast episode's audio bill: MAI, Lyria and transcript vs Sume
A 5,000-character intro, one music track and a 60-minute transcript cost $0.695 on MAI Flash, Lyria 3.5 and MAI-Transcribe, and $0.9625 on Sume.
- LiveTranslate 2.3 s latency vs STT plus TTS
Qwen3.8-LiveTranslate cut average lagging from 2.8 to 2.3 seconds. Sume has no live interpreter, but stt_create and tts_create can build an offline dub.
- Recast, Omni edit, Genjutsu or a new shot: which ad variant method?
Four ways to make an ad variant on Sume, limits side by side: Recast swaps people, Omni edit changes by prompt, Genjutsu moves motion, a new shot starts over.
- Recast one holiday ad for four markets: fal vs Sume cost
Four 15-second Recast runs cost $18.00 at fal's $0.30 a second and $22.50 on Sume at $0.375 a second, both at 768p. See the math and what the gap buys.
Written by Sume