v0.3.0 — Batch Transcription Support¶
This release adds batch transcription support for pre-recorded audio files and URLs, complementing the existing real-time streaming transcription.
Highlights¶
- Batch transcription support for pre-recorded audio files and URLs
Added¶
Transcribemethod for batch transcription of audio bytes (d64c4ea)TranscribeFilemethod for batch transcription of local audio files (d64c4ea)TranscribeURLmethod for batch transcription of remote audio URLs (d64c4ea)- Pre-recorded transcription conversion helpers for Deepgram responses (
3fc08dd)
Tests¶
TranscribeFileandTranscribeURLconformance tests with Ferris Bueller test audio (35c593e)
Dependencies¶
- Bump
github.com/agentplexus/omnivoicefrom v0.3.0 to v0.4.1 (35c593e)
Documentation¶
- README updated with batch transcription examples (
fe09c20)
Quick Start¶
import (
deepgramstt "github.com/agentplexus/omnivoice-deepgram/omnivoice/stt"
"github.com/agentplexus/omnivoice/stt"
)
// Create STT provider
provider, err := deepgramstt.New(deepgramstt.WithAPIKey("your-api-key"))
if err != nil {
log.Fatal(err)
}
// Transcribe a local file
result, err := provider.TranscribeFile(ctx, "audio.mp3", stt.TranscriptionConfig{
Language: "en",
})
if err != nil {
log.Fatal(err)
}
fmt.Println(result.Text)
// Transcribe from URL
result, err = provider.TranscribeURL(ctx, "https://example.com/audio.mp3", stt.TranscriptionConfig{
Language: "en",
})
See the README for more examples.