🔥 Qwen/Qwen3-32B
High-performance 32B parameter LLM. Excellent for reasoning, coding, and multilingual tasks.
Input: $0.15/M tokensOutput: $0.44/M tokens
33.54M runs in 7 days
Industry-leading speech recognition across 100+ languages with near-human accuracy.
Parameters
1.55B
Context
30 seconds (auto-chunked)
Organization
OpenAI
Start using Whisper Large V3 in minutes. VoltageGPU provides an OpenAI-compatible API — just change the base_url.
from openai import OpenAI
client = OpenAI(
base_url="https://api.voltagegpu.com/v1",
api_key="YOUR_VOLTAGE_API_KEY"
)
# Transcribe audio file
with open("meeting_recording.mp3", "rb") as audio_file:
transcript = client.audio.transcriptions.create(
model="whisper-large-v3",
file=audio_file,
response_format="verbose_json",
timestamp_granularities=["word", "segment"]
)
print(transcript.text)
# Access word-level timestamps
for word in transcript.words:
print(f"[{word.start:.2f}s] {word.word}")curl -X POST https://api.voltagegpu.com/v1/audio/transcriptions \
-H "Authorization: Bearer YOUR_VOLTAGE_API_KEY" \
-F file="@meeting_recording.mp3" \
-F model="whisper-large-v3" \
-F response_format="verbose_json" \
-F timestamp_granularities[]="word"| Component | Price | Unit |
|---|---|---|
| Speech Recognition | $0.006 | per minute of audio |
New accounts receive $5 free credit. No credit card required to start.
Whisper Large V3 achieves a word error rate (WER) below 5% on English and under 10% on many other languages. It supports 100+ languages for transcription and automatic language detection. Features include word-level timestamps, speaker diarization hints, punctuation restoration, and translation from any supported language to English. The model handles audio formats including MP3, WAV, FLAC, M4A, OGG, and WebM.
Whisper Large V3 is OpenAI's most capable automatic speech recognition (ASR) model, supporting transcription and translation across 100+ languages. With 1.55 billion parameters, it achieves near-human accuracy on English speech and strong performance on dozens of other languages. Whisper V3 handles accented speech, background noise, technical jargon, and multiple speakers with remarkable accuracy. It supports automatic language detection, timestamps at word and segment level, and audio-to-text translation from any supported language to English.
Transcribe meetings, interviews, and conferences with accurate speaker identification.
Automatically transcribe podcasts for searchable archives, show notes, and accessibility.
Generate accurate subtitles and closed captions for video content in multiple languages.
Build voice-controlled applications with reliable speech-to-text conversion.
Transcribe professional dictation with support for domain-specific terminology.
https://api.voltagegpu.com/v1/audio/transcriptions| Authorization | Bearer YOUR_VOLTAGE_API_KEY | Required |
| Content-Type | application/json | Required |
openai/whisper-large-v3Use this value as the model parameter in your API requests.
curl -X POST https://api.voltagegpu.com/v1/audio/transcriptions \
-H "Authorization: Bearer YOUR_VOLTAGE_API_KEY" \
-F file="@meeting_recording.mp3" \
-F model="whisper-large-v3" \
-F response_format="verbose_json" \
-F timestamp_granularities[]="word"Great price-performance for smaller models with 24GB VRAM.
Enterprise-grade GPU for production inference at scale.
Access this model and 140+ others through our OpenAI-compatible API.
Compare GPU cloud pricing and model hosting features.
View GPU compute and AI inference pricing with no hidden fees.
Deploy a GPU pod in under 60 seconds to run models locally.
Whisper V3 supports transcription in 100+ languages including English, Chinese, Spanish, French, German, Japanese, Korean, Arabic, Hindi, Portuguese, Russian, and many more. It can also translate speech from any supported language into English text.
Whisper V3 achieves below 5% word error rate (WER) on clean English speech, comparable to professional human transcriptionists. On other languages, accuracy varies but typically achieves 5-15% WER for well-represented languages.
Whisper V3 accepts MP3, WAV, FLAC, M4A, OGG, WebM, and other common audio formats. The maximum file size is typically 25MB, with longer files automatically chunked for processing.
Whisper V3 transcribes all speech in an audio file but does not natively distinguish between speakers. For speaker diarization (identifying who said what), you can combine Whisper with a separate diarization model.
At $0.006 per minute, transcribing 1 hour of audio costs $0.36. This makes Whisper V3 via VoltageGPU one of the most affordable transcription options available, significantly cheaper than human transcription ($1-3/minute) and most competing APIs.
Get $5 free credit when you sign up. No credit card required. Deploy in under 30 seconds with our OpenAI-compatible API.