ref #10 : quick-and-dirty attempt for real-time audio transciption

- Processes input in chunks of 3 seconds.
- Padding audio with silence
- Uses 1 second audio from previous pass
- No text context
This commit is contained in:
Georgi Gerganov 2022-10-02 17:55:45 +03:00
parent 77d929f603
commit b6bf906730
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
3 changed files with 2517 additions and 0 deletions

1
.gitignore vendored
View File

@ -1,3 +1,4 @@
sync.sh
main
stream
*.o

View File

@ -1,3 +1,5 @@
CC_SDL=`sdl2-config --cflags --libs`
main: ggml.o main.o
g++ -pthread -o main ggml.o main.o
./main -h
@ -8,6 +10,9 @@ ggml.o: ggml.c ggml.h
main.o: main.cpp ggml.h
g++ -pthread -O3 -std=c++11 -c main.cpp
stream: stream.cpp
g++ -pthread -O3 -std=c++11 -o stream stream.cpp ggml.o $(CC_SDL)
# clean up the directory
clean:
rm -f *.o main

2511
stream.cpp Normal file

File diff suppressed because it is too large Load Diff