frontmatter in skills files
This commit is contained in:
@@ -16,6 +16,18 @@ whisper.cpp is a C/C++ port of OpenAI's Whisper speech recognition model. The se
|
||||
|
||||
## Endpoints
|
||||
|
||||
### GET /health
|
||||
|
||||
Returns server status. Use to verify the server is running before making requests.
|
||||
|
||||
```bash
|
||||
curl -s http://macmini:8178/health
|
||||
```
|
||||
|
||||
```json
|
||||
{"status":"ok"}
|
||||
```
|
||||
|
||||
### POST /inference
|
||||
|
||||
Transcribes an audio file. Accepts `multipart/form-data`.
|
||||
@@ -167,6 +179,76 @@ curl -s http://macmini:8178/load \
|
||||
|
||||
The server accepts at least WAV (16-bit PCM) and MP3 files directly. If the server was started with `--convert`, it can use ffmpeg to handle additional formats (ogg, flac, m4a, etc.).
|
||||
|
||||
## Common Recipes
|
||||
|
||||
### Transcribe to plain text
|
||||
|
||||
```bash
|
||||
curl -s http://macmini:8178/inference \
|
||||
-F file="@audio.mp3" \
|
||||
-F temperature="0.0" \
|
||||
-F temperature_inc="0.2" \
|
||||
-F response_format="text"
|
||||
```
|
||||
|
||||
### Transcribe non-English audio
|
||||
|
||||
```bash
|
||||
curl -s http://macmini:8178/inference \
|
||||
-F file="@audio.mp3" \
|
||||
-F language="pt" \
|
||||
-F temperature="0.0" \
|
||||
-F temperature_inc="0.2" \
|
||||
-F response_format="json"
|
||||
```
|
||||
|
||||
### Translate foreign audio to English
|
||||
|
||||
```bash
|
||||
curl -s http://macmini:8178/inference \
|
||||
-F file="@audio.mp3" \
|
||||
-F language="auto" \
|
||||
-F translate="true" \
|
||||
-F temperature="0.0" \
|
||||
-F temperature_inc="0.2" \
|
||||
-F response_format="json"
|
||||
```
|
||||
|
||||
### Generate SRT subtitles
|
||||
|
||||
```bash
|
||||
curl -s http://macmini:8178/inference \
|
||||
-F file="@audio.mp3" \
|
||||
-F temperature="0.0" \
|
||||
-F temperature_inc="0.2" \
|
||||
-F response_format="srt" \
|
||||
> subtitles.srt
|
||||
```
|
||||
|
||||
### Transcribe with VAD (skip silence)
|
||||
|
||||
```bash
|
||||
curl -s http://macmini:8178/inference \
|
||||
-F file="@audio.mp3" \
|
||||
-F temperature="0.0" \
|
||||
-F temperature_inc="0.2" \
|
||||
-F vad="true" \
|
||||
-F response_format="json"
|
||||
```
|
||||
|
||||
### Transcribe with vocabulary hints
|
||||
|
||||
```bash
|
||||
curl -s http://macmini:8178/inference \
|
||||
-F file="@audio.mp3" \
|
||||
-F temperature="0.0" \
|
||||
-F temperature_inc="0.2" \
|
||||
-F prompt="Kubernetes, kubectl, etcd, gRPC" \
|
||||
-F response_format="json"
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Source
|
||||
|
||||
- Repository: https://github.com/ggml-org/whisper.cpp
|
||||
|
||||
File diff suppressed because one or more lines are too long
Reference in New Issue
Block a user