thread/hyprvoice
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

add gpt-4o-transcribe models to openai provider

Browse Source
This commit is contained in:
leonardotrapani
2026-02-01 00:44:10 +01:00
parent 16ef8bcf0a
commit 9719df7d69
4 changed files with 37 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files
progress.txt
+8 -5
View File
@@ -10,13 +10,16 @@ Started: Sun Feb 1 12:22:47 AM CET 2026
- Full 57 language list from OpenAI Whisper
- All tests passing, typecheck passes
### Task 2: Add language list and provider-specific mappings
- Added `ToProviderFormat(code string, providerName string) string` function
- Handles auto-detect: whisper-cpp uses "auto", others use ""
- Deepgram locale mappings: en->en-US, pt->pt-BR, zh->zh-CN
- Added comprehensive test coverage for all provider formats
### Task 8: Migrate OpenAI provider to new Model structure
- Added gpt-4o-transcribe and gpt-4o-mini-transcribe transcription models
- Removed gpt-4-turbo and gpt-3.5-turbo LLM models (not in PRD)
- OpenAI now has 5 models: 3 transcription (whisper-1, gpt-4o-transcribe, gpt-4o-mini-transcribe) + 2 LLM (gpt-4o-mini, gpt-4o)
- All models have AdapterType='openai', Endpoint with BaseURL='https://api.openai.com'
- All models use language.AllLanguageCodes() for SupportedLanguages (57 languages)
- Updated provider_test.go to expect 3 transcription and 2 LLM models
- All tests passing, typecheck passes
### Task 3: Create Model type with full metadata
- Created `internal/provider/model.go`
- ModelType enum: Transcription, LLM