Chirp Transcription

Developer documentation

Google Cloud Speech-to-Text Chirp for speech recognition and transcription.

Model Reference

Audio and speech models

Text-to-speech, GPT audio chat, realtime sessions, transcription-oriented catalog rows, and voice settings. Endpoint: https://omixa.cloud/api/v1/audio

Chirp Transcription

chirp

Google Cloud Speech-to-Text Chirp for speech recognition and transcription.

声音的

Audio per minute $0.016000

Minimum hold $0.010000

Integration reference

Connect Chirp Transcription

Use Omixa's unified endpoint and your workspace API key. Provider routing, billing, failover, and usage records are handled by Omixa.

POST https://omixa.cloud/api/v1/audio

Provider: google
Endpoint type: google_cloud_speech_to_text

Request schema

Request fields

Only send options supported by this model. Required fields and accepted values are listed below.

Field	类型	Required	Accepted values	描述
`model`	string	Yes	chirp	Use `chirp`. Omixa resolves the active provider route and failover key automatically.
`task`	string	No	transcription	Speech workflow supported by this model.
`file_data`	base64\|string	Yes	Any valid value	Source speech audio as raw base64 or an audio data URL.
`filename`	string	Yes	Any valid value	Original filename including a supported audio extension.
`mime_type`	string	Yes	Any valid value	Audio MIME type such as `audio/wav`, `audio/mpeg`, or `audio/flac`.
`language`	string	No	Any valid value	Source BCP-47 language code. Chirp 3 can use `auto` for automatic language recognition.
`response_format`	string	No	json	Normalized transcript response format.

Ready to send

Payload and response

Start with this model-safe payload and expect the normalized Omixa response shape shown beside it.

Example JSON payload

{
    "model": "chirp",
    "task": "transcription",
    "file_data": "data:audio/wav;base64,<base64-audio>",
    "filename": "speech.wav",
    "mime_type": "audio/wav",
    "language": "en-US",
    "response_format": "json"
}

Response shape

{
    "object": "audio.transcription",
    "text": "Transcribed speech...",
    "segments": [],
    "words": [],
    "usage": {
        "estimated_audio_minutes": "0.100000"
    }
}

Language examples

Copy-ready integration code

Replace the example API key with a workspace key and keep model-specific fields unchanged unless the table above marks them optional.

curl -X POST https://omixa.cloud/api/v1/audio \
  -H "Authorization: Bearer omx_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "chirp",
    "task": "transcription",
    "file_data": "data:audio/wav;base64,<base64-audio>",
    "filename": "speech.wav",
    "mime_type": "audio/wav",
    "language": "en-US",
    "response_format": "json"
}'

const response = await fetch('https://omixa.cloud/api/v1/audio', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer omx_live_xxx',
    'Content-Type': 'application/json'
  },
  body: "{\n    \"model\": \"chirp\",\n    \"task\": \"transcription\",\n    \"file_data\": \"data:audio/wav;base64,<base64-audio>\",\n    \"filename\": \"speech.wav\",\n    \"mime_type\": \"audio/wav\",\n    \"language\": \"en-US\",\n    \"response_format\": \"json\"\n}"
});
const data = await response.json();

import requests

response = requests.post(
    'https://omixa.cloud/api/v1/audio',
    headers={'Authorization': 'Bearer omx_live_xxx', 'Content-Type': 'application/json'},
    json={
    "model": "chirp",
    "task": "transcription",
    "file_data": "data:audio/wav;base64,<base64-audio>",
    "filename": "speech.wav",
    "mime_type": "audio/wav",
    "language": "en-US",
    "response_format": "json"
}
)
print(response.json())

$ch = curl_init('https://omixa.cloud/api/v1/audio');
curl_setopt_array($ch, [
    CURLOPT_POST => true,
    CURLOPT_HTTPHEADER => ['Authorization: Bearer omx_live_xxx', 'Content-Type: application/json'],
    CURLOPT_POSTFIELDS => '{
    "model": "chirp",
    "task": "transcription",
    "file_data": "data:audio/wav;base64,<base64-audio>",
    "filename": "speech.wav",
    "mime_type": "audio/wav",
    "language": "en-US",
    "response_format": "json"
}',
    CURLOPT_RETURNTRANSFER => true,
]);
$response = curl_exec($ch);

using var client = new HttpClient();
client.DefaultRequestHeaders.Authorization = new System.Net.Http.Headers.AuthenticationHeaderValue("Bearer", "omx_live_xxx");
var json = @"{
    ""model"": ""chirp"",
    ""task"": ""transcription"",
    ""file_data"": ""data:audio/wav;base64,<base64-audio>"",
    ""filename"": ""speech.wav"",
    ""mime_type"": ""audio/wav"",
    ""language"": ""en-US"",
    ""response_format"": ""json""
}";
var response = await client.PostAsync("https://omixa.cloud/api/v1/audio", new StringContent(json, System.Text.Encoding.UTF8, "application/json"));
var body = await response.Content.ReadAsStringAsync();

payload := []byte(`{
    "model": "chirp",
    "task": "transcription",
    "file_data": "data:audio/wav;base64,<base64-audio>",
    "filename": "speech.wav",
    "mime_type": "audio/wav",
    "language": "en-US",
    "response_format": "json"
}`)
req, _ := http.NewRequest("POST", "https://omixa.cloud/api/v1/audio", bytes.NewReader(payload))
req.Header.Set("Authorization", "Bearer omx_live_xxx")
req.Header.Set("Content-Type", "application/json")
resp, err := http.DefaultClient.Do(req)

Production checklist

Operational notes

Authenticate with `Authorization: Bearer omx_live_xxx`.
Omixa handles provider keys, routing, billing, failover, and usage recording behind this endpoint.
Google Vertex requests use active Vertex accounts configured by the admin; Omixa retries the next healthy account when a route fails before output starts.