Transcribe Video with Sapat and Deepgram - #326
Closed
honanevan wants to merge 4 commits into
Closed
Conversation
Author
|
Withdrawing this draft after Daytona support confirmed today that this content programme is inactive and relates to the former 2024 product. No reward or acceptance is claimed. The companion Sapat implementation and validation remain available in nibzard/sapat#79. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
/claim #13
Guide and implementation companion for the AI Transcription Tool bounty. Adds a long-form guide, original pipeline illustration, HTTP contract-test definition and author profile. Companion provider PR: nibzard/sapat#79
Executed inside Daytona on 8 September 2026 using snapshot daytonaio/sandbox:0.8.0, EU region, 1 vCPU/1 GiB RAM/3 GiB disk, Python 3.14.4, Git 2.53.0 and FFmpeg 7.1.3:
The guide now documents the actual dashboard, sandbox and terminal steps. Markdownlint passes. Separate Linux CI on Python 3.10/3.12 also passes: https://github.com/honanevan/sapat/actions/runs/34167025609 . Local Windows results were 189 pass/one existing WhisperX path failure; the full Daytona run passes.
Live validation update (8 September): the full CLI successfully converted an original synthetic MP4, called Deepgram nova-3 with language en using free account credit, and saved a transcript matching the recording's source text on Windows. This is one end-to-end example, not an accuracy benchmark. The identical provider revision was tested.
The live call inside Daytona converted the video but failed with ConnectionResetError(104) during POST to api.deepgram.com/v1/listen. Daytona's published Tier 1/2 network allowlist does not include Deepgram. Support is reviewing ticket #18120, including the question of a supported no-cost endpoint-access option and current programme eligibility. This is not claimed as a successful live Daytona transcription.
Remaining before publication: resolve Daytona network access and produce the requested short Loom demo. This remains a draft pending those requirements and editorial review.
Prepared with AI assistance and fact-checked against current source and official documentation. Author attribution uses the contributor's GitHub handle. No payment is claimed as received.