Premiere 26.5 Match Source audio vs audio detach channels and rate

Premiere 26.5 adds a Match Source option for audio channels and sample rate, with warnings when a downmix is needed. If you want the same outcome without an editor, Sume's POST /v1/audio-detach pulls the audio out of a Sume-hosted video as WAV or MP3 and lets you set channels to source or mono and sample_rate to 16000, 44100 or 48000. For speech-to-text, 16000 Hz mono is the documented shape.

What each tool controls

Premiere's Match Source concerns an export or sequence setting: the application picks channel layout and sample rate from the source and warns you if it has to mix down. Audio detach is a separate step that writes a standalone audio file. format is wav (default, pcm_s16le) or mp3 at 128 kbps. An optional range limits the extract. When sample_rate is omitted, the result reports it as null and the value is inherited from the source.

A speech-ready extract

The call below produces a mono 16 kHz WAV, which is what you hand to speech-to-text or a dubbing step. The route costs $0.01 per job; confirm it in GET /v1/catalog. If the file has no audio track the job fails with detach_source_has_no_audio, so probe with a video-inspect call first when you are unsure.

curl -X POST https://api.sume.com/v1/audio-detach \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: detach-stt-001" \
  -d '{"video_url":"https://media.sume.com/artifacts/artf_demo/talk.mp4","format":"wav","channels":"mono","sample_rate":16000}'

Honest limits

• Source video is limited to 1800 seconds and the extracted output to 900 seconds, so longer programs need ranges.
• The detach route does not warn about downmixing the way an editor does; choose channels deliberately.
• Adobe's post does not detail the downmix algorithm, so this post does not claim equivalence.