metadata
license: mit
pipeline_tag: automatic-speech-recognition
OpenAI's Whisper models converted to ggml format
Model | Disk | Mem | SHA |
---|---|---|---|
tiny | 75 MB | ~390 MB | bd577a113a864445d4c299885e0cb97d4ba92b5f |
tiny.en | 75 MB | ~390 MB | c78c86eb1a8faa21b369bcd33207cc90d64ae9df |
base | 142 MB | ~500 MB | 465707469ff3a37a2b9b8d8f89f2f99de7299dac |
base.en | 142 MB | ~500 MB | 137c40403d78fd54d454da0f9bd998f78703390c |
small | 466 MB | ~1.0 GB | 55356645c2b361a969dfd0ef2c5a50d530afd8d5 |
small.en | 466 MB | ~1.0 GB | db8a495a91d927739e50b3fc1cc4c6b8f6c2d022 |
medium | 1.5 GB | ~2.6 GB | fd9727b6e1217c2f614f9b698455c4ffd82463b4 |
medium.en | 1.5 GB | ~2.6 GB | 8c30f0e44ce9560643ebd10bbe50cd20eafd3723 |
large-v1 | 2.9 GB | ~4.7 GB | b1caaf735c4cc1429223d5a74f0f4d0b9b59a299 |
large-v2 | 2.9 GB | ~4.7 GB | 0f4c8e34f21cf1a914c59d8b3ce882345ad349d6 |
large | 2.9 GB | ~4.7 GB | ad82bf6a9043ceed055076d0fd39f5f186ff8062 |
note: large
corresponds to the latest Large v3 model
For more information, visit: