Audio (Speech & Transcription)

Text-to-Speech

POST /v1/audio/speech

curl https://balitechsolution.com/v1/audio/speech \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "input": "Halo dari Bali!",
    "voice": "voice_id",
    "response_format": "mp3"
  }' --output speech.mp3

Parameter	Type	Default	Description
input	string	required	Teks yang mau diucapkan
voice	string	default voice	ID suara, lihat Voices
model	string	eleven_flash_v2_5	Model TTS (ElevenLabs)
response_format	string	mp3	Format audio output

Response berupa binary audio (audio/mpeg untuk mp3).

Speech-to-Text

POST /v1/audio/transcriptions

Kirim file audio sebagai multipart/form-data, sama seperti OpenAI Whisper API.

curl https://balitechsolution.com/v1/audio/transcriptions \
  -H "Authorization: Bearer sk-xxx" \
  -F [email protected]

Response JSON berisi hasil transkripsi (diproses via Deepgram nova-3, smart_format aktif).

Voices

GET /v1/voices

Mengembalikan daftar suara yang tersedia untuk text-to-speech.