Audio, speech & music models

Entraînement

Objectives vary: transcription uses sequence likelihood, representation learning may mask audio, codec generators predict discrete audio tokens, and diffusion systems learn denoising in waveform, spectrogram, or latent space.

Exécution de l’inférence

ASR usually encodes audio then decodes text; generation may predict codec tokens autoregressively or iteratively denoise. Streaming constraints can require chunked encoders and bounded look-ahead.

Atouts

• •Speech transcription and translation
• •Natural speech, sound, and music synthesis
• •Audio embeddings support search, moderation, and classification

Compromis

• •Accents, noise, overlap, code-switching, and domain vocabulary shift quality
• •Long-form generation needs temporal structure and consistency
• •Voice identity, consent, watermarking, and copyright need explicit governance

À utiliser lorsque

• The modality itself carries needed information
• You have task-, language-, and environment-specific evaluations
• Latency and streaming requirements are designed up front

À éviter ou remettre en question lorsque

• A transcript already contains all relevant information
• Identity cloning lacks informed consent
• One benchmark is being used as proof across languages and acoustic conditions