Audio

Speech to text and text to speech.

Base URL https://api.siliconnetwork.io

POST/v1/audio/transcriptions
POST/v1/audio/speech

Audio Transcriptions

Multipart upload. response_format is json, text, verbose_json (segments and language), srt, or vtt. Billed per second of audio.

silicon - zsh

curl -X POST https://api.siliconnetwork.io/v1/audio/transcriptions \
  -H "Authorization: Bearer sn_live_..." \
  -F "file=@meeting.wav" \
  -F "model=whisper-large-v3-turbo" \
  -F "response_format=verbose_json"

transcribe.py

from openai import OpenAI

client = OpenAI(base_url="https://api.siliconnetwork.io/v1", api_key="sn_live_...")
with open("meeting.wav", "rb") as fh:
    out = client.audio.transcriptions.create(model="whisper-large-v3-turbo", file=fh)
print(out.text)

response

{
  "text": "Thanks everyone, let's start with the roadmap.",
  "language": "en",
  "duration": 4.2,
  "segments": [{ "id": 0, "start": 0.0, "end": 4.2, "text": "Thanks everyone, let's start with the roadmap." }],
  "usage": { "type": "duration", "seconds": 5 }
}

Responses: 200 Successful Response, 422 Validation Error

POST/v1/audio/speech

Audio Speech

Returns audio bytes; the Content-Type names the format (wav and pcm always, mp3 when the node has an encoder). Billed per character of input; the count is echoed in X-Silicon-Usage-Units.

silicon - zsh

curl -X POST https://api.siliconnetwork.io/v1/audio/speech \
  -H "Authorization: Bearer sn_live_..." \
  -H "Content-Type: application/json" \
  -d '{
  "model": "kokoro-82m",
  "input": "Welcome to Silicon Network.",
  "voice": "af_heart",
  "response_format": "wav"
}'

request.py

from siliconnetwork import SiliconNetwork

client = SiliconNetwork()
data, meta = client.request("POST", "/v1/audio/speech", json={
  "model": "kokoro-82m",
  "input": "Welcome to Silicon Network.",
  "voice": "af_heart",
  "response_format": "wav"
})
print(data, meta.request_id)

speak.py

from openai import OpenAI

client = OpenAI(base_url="https://api.siliconnetwork.io/v1", api_key="sn_live_...")
with client.audio.speech.with_streaming_response.create(
    model="kokoro-82m", voice="af_heart", input="Welcome to Silicon Network.", response_format="wav"
) as resp:
    resp.stream_to_file("welcome.wav")

Responses: 200 Successful Response, 422 Validation Error

Silicon Network›Docs›API›Audio37 items