Whisper is a pre-trained model for automatic speech recognition (ASR) and speech translation. Trained on 680k hours of labelled data, Whisper models demonstrate a strong ability to generalize to many datasets and domains without the need for fine-tuning. This is the English-only version of the Whisper Tiny model which was trained on the task of speech recognition.
Whisper-Tiny-EN (openai/whisper-tiny-en) is a audio-stt model from OpenAI, released 2024-04-22. Pricing via AIgateway: $0.0040 per minute. Call it via https://api.aigateway.sh/v1/audio/transcriptions — set model="openai/whisper-tiny-en".
curl https://api.aigateway.sh/v1/audio/transcriptions \
-H "Authorization: Bearer $AIGATEWAY_API_KEY" \
-F model="openai/whisper-tiny-en" \
-F file="@audio.mp3"# multipart/form-data — use curl -F or SDK file upload model="openai/whisper-tiny-en" file=@audio.mp3 response_format=json # or "verbose_json", "text", "srt", "vtt" language=en # optional
{
"text": "Hello from AIgateway.",
"language": "en",
"duration": 1.82
}from openai import OpenAI
client = OpenAI(base_url="https://api.aigateway.sh/v1", api_key="sk-aig-...")
with open("audio.mp3", "rb") as f:
r = client.audio.transcriptions.create(model="openai/whisper-tiny-en", file=f)
print(r.text)