feat: AI text to speech for proposal body #244
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files📢 Thoughts on this report? Let us know! |
|
For reviewers, check the approved sister PR #240 for an idea on the class structure and flow |
| AI_SUMMARY_SUBDIR=ai-summary | ||
| AI_TTS_SUBDIR=ai-tts |
There was a problem hiding this comment.
Could we also add default values for these, if someone forget. also not changing in .env.test
There was a problem hiding this comment.
Default value is undefined, as defined here
snapshot-sidekick/src/helpers/utils.ts
Line 41 in 61a39b5
|
Somehow i always get Any idea? |
Are you sending a POST request ? |
| } | ||
| }); | ||
|
|
||
| router.post('/ai/tts/:id', async (req, res) => { |
There was a problem hiding this comment.
Any reason to use post here instead of get and for summery when we are not using any input from body
There was a problem hiding this comment.
Very basic bot protection, to avoid link prefetch or any other script/tools calling this API endpoint
AI text to speech
Generate a .mp3 audio file through OpenAI API from the proposal's body
NOTE:
Test
OPENAI_API_KEYenv var set in your.env.env, setSTORAGE_ENGINEtofile, to save the mp3 file in/tmplocally, for easier accesshttp://localhost:3005/api/ai/tts/[PROPOSAL-ID](choose a proposal with a body length between 500 and 4096 chars)