ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-06-17 06:18:08 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

server : implement "verbose_json" format with token details (#1781)

Browse Source 
* examples/server: implement "verbose_json" format with token details.

This is intended to mirror the format of openai's Python
whisper.transcribe() return values.

* server: don't write WAV to a temporary file if not converting

* server: use std::lock_guard instead of manual lock/unlock
This commit is contained in:
Ryan Hitchman
2024-01-18 13:58:42 -07:00
committed by GitHub
parent fb466b3417
commit c0329acde8
3 changed files with 75 additions and 31 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
examples/common.h
View File

@ -136,6 +136,7 @@ gpt_vocab::id gpt_sample_top_k_top_p_repeat(
//
// Read WAV audio file and store the PCM data into pcmf32
// fname can be a buffer of WAV data instead of a filename
// The sample rate of the audio must be equal to COMMON_SAMPLE_RATE
// If stereo flag is set and the audio has 2 channels, the pcmf32s will contain 2 channel PCM
bool read_wav(