Whisper Transcribe - Audio Transcription Example

This example demonstrates audio transcription using OpenAI Whisper models.

Run command:

cargo run --example whisper_transcribe --release --features inference

Topics covered:

• Whisper model architecture (encoder-decoder)
• Audio preprocessing (mel spectrogram)
• Beam search decoding
• Timestamp extraction
• Multi-language transcription

Supported models:

• whisper-tiny (39M parameters)
• whisper-base (74M parameters)
• whisper-small (244M parameters)
• whisper-medium (769M parameters)

See also:

• Audio Mel Spectrogram
• Examples Reference