| --- |
| license: mit |
| language: |
| - af |
| base_model: openai/whisper-medium |
| library_name: ctranslate2 |
| pipeline_tag: automatic-speech-recognition |
| tags: |
| - whisper |
| - fluister |
| - afrikaans |
| - south-africa |
| - faster-whisper |
| - ctranslate2 |
| --- |
| |
| # Fluister (medium) |
|
|
| **Fluister** is an Afrikaans-optimised Whisper. ("Fluister" is Afrikaans for "to whisper".) It is a |
| fine-tune of OpenAI Whisper (openai/whisper-medium), merged into the base weights and converted to CTranslate2 |
| (int8) for use with faster-whisper. By DigiPhyte (Pty) Ltd, South Africa. |
|
|
| On real South African Afrikaans and Afrikaans/English code-switched audio it produces clean |
| Afrikaans where stock Whisper drifts to Dutch spellings ("gebou" not "gebouw", "mense" not |
| "mensen", "eintlik" not "eindelijk"), while keeping English code-switching intact. |
|
|
| ## Use (faster-whisper) |
|
|
| ```python |
| from faster_whisper import WhisperModel |
| model = WhisperModel("digiphyte/fluister-medium", device="cuda", compute_type="int8_float16") # CPU: device="cpu", compute_type="int8" |
| segments, info = model.transcribe("audio.wav", language="af", beam_size=5) |
| for s in segments: |
| print(s.text) |
| ``` |
|
|
| Pass `language="af"`; the Fluister models are tuned for Afrikaans and SA English and should be told |
| the language rather than relying on auto-detect. |
|
|
| ## Licence and attribution |
|
|
| MIT (see `LICENSE`). This is a derivative work; the base model (OpenAI Whisper, Apache-2.0) and the training data (`andreoosthuizen/afrikaans-30s`, CC-BY-4.0) (this size was fine-tuned by DigiPhyte directly on that dataset) are credited in `NOTICE`. |
|
|
