whatsapp-chatgpt icon indicating copy to clipboard operation
whatsapp-chatgpt copied to clipboard

Using other voice languages like Japan/korea beside English, crashes the whatsapp.

Open vcoopers opened this issue 2 years ago • 4 comments

Somehow if I speak in other foreign languages, it crashes WhatsApp.

◇ [Transcription] Transcribing audio with "local" mode... C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py:79: UserWarning: FP16 is not supported on CPU; using FP32 instead warnings.warn("FP16 is not supported on CPU; using FP32 instead") Traceback (most recent call last): File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\Scripts\whisper-script.py", line 33, in sys.exit(load_entry_point('openai-whisper==20230124', 'console_scripts', 'whisper')()) File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 314, in cli result = transcribe(model, audio_path, temperature=temperature, **args) File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 230, in transcribe add_segment( File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 170, in add_segment print(make_safe(f"[{format_timestamp(start)} --> {format_timestamp(end)}] {text}")) File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\encodings\cp1252.py", line 19, in encode return codecs.charmap_encode(input,self.errors,encoding_table)[0] UnicodeEncodeError: 'charmap' codec can't encode characters in position 26-30: character maps to node:internal/errors:867 const err = new Error(message); ^

Error: Command failed: whisper C:\Users\vcoop\AppData\Local\Temp\6fabd890-156c-475b-a2e2-9838a2c19884.wav C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py:79: UserWarning: FP16 is not supported on CPU; using FP32 instead warnings.warn("FP16 is not supported on CPU; using FP32 instead") Traceback (most recent call last): File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\Scripts\whisper-script.py", line 33, in sys.exit(load_entry_point('openai-whisper==20230124', 'console_scripts', 'whisper')()) File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 314, in cli result = transcribe(model, audio_path, temperature=temperature, **args) File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 230, in transcribe add_segment( File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 170, in add_segment print(make_safe(f"[{format_timestamp(start)} --> {format_timestamp(end)}] {text}")) File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\encodings\cp1252.py", line 19, in encode return codecs.charmap_encode(input,self.errors,encoding_table)[0] UnicodeEncodeError: 'charmap' codec can't encode characters in position 26-30: character maps to

at checkExecSyncError (node:child_process:885:11)
at Proxy.execSync (node:child_process:957:15)
at Module.transcribeAudioLocal (E:\OneDrive\Documents\GitHub\whatsapp-chatgpt\src\providers\whisper-local.ts:15:40)
at Module.handleIncomingMessage (E:\OneDrive\Documents\GitHub\whatsapp-chatgpt\src\handlers\message.ts:33:62)
at processTicksAndRejections (node:internal/process/task_queues:95:5)
at Client.<anonymous> (E:\OneDrive\Documents\GitHub\whatsapp-chatgpt\src\index.ts:46:5) {

status: 1, signal: null, output: [ null, 'Detecting language using up to the first 30 seconds. Use --language to specify the language\r\n' + 'Detected language: Japanese\r\n', 'C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py:79: UserWarning: FP16 is not supported on CPU; using FP32 instead\r\n' + ' warnings.warn("FP16 is not supported on CPU; using FP32 instead")\r\n' + 'Traceback (most recent call last):\r\n' + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\Scripts\whisper-script.py", line 33, in \r\n' + " sys.exit(load_entry_point('openai-whisper==20230124', 'console_scripts', 'whisper')())\r\n" + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 314, in cli\r\n' + ' result = transcribe(model, audio_path, temperature=temperature, **args)\r\n' + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 230, in transcribe\r\n' + ' add_segment(\r\n' + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 170, in add_segment\r\n' + ' print(make_safe(f"[{format_timestamp(start)} --> {format_timestamp(end)}] {text}"))\r\n' + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\encodings\cp1252.py", line 19, in encode\r\n' + ' return codecs.charmap_encode(input,self.errors,encoding_table)[0]\r\n' + "UnicodeEncodeError: 'charmap' codec can't encode characters in position 26-30: character maps to \r\n" ], pid: 42600, stdout: 'Detecting language using up to the first 30 seconds. Use --language to specify the language\r\n' + 'Detected language: Japanese\r\n', stderr: 'C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py:79: UserWarning: FP16 is not supported on CPU; using FP32 instead\r\n' +
' warnings.warn("FP16 is not supported on CPU; using FP32 instead")\r\n' + 'Traceback (most recent call last):\r\n' + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\Scripts\whisper-script.py", line 33, in \r\n' + " sys.exit(load_entry_point('openai-whisper==20230124', 'console_scripts', 'whisper')())\r\n" + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 314, in cli\r\n' + ' result = transcribe(model, audio_path, temperature=temperature, **args)\r\n' + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 230, in transcribe\r\n' + ' add_segment(\r\n' + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\site-packages\whisper\transcribe.py", line 170, in add_segment\r\n' + ' print(make_safe(f"[{format_timestamp(start)} --> {format_timestamp(end)}] {text}"))\r\n' + ' File "C:\Users\vcoop\AppData\Local\Programs\Python\Python310\lib\encodings\cp1252.py", line 19, in encode\r\n' + ' return codecs.charmap_encode(input,self.errors,encoding_table)[0]\r\n' + "UnicodeEncodeError: 'charmap' codec can't encode characters in position 26-30: character maps to \r\n"

vcoopers avatar Feb 24 '23 06:02 vcoopers

Currently speech only supports english, we have plans to implement other langauges in the future :)

connorv001 avatar Feb 24 '23 06:02 connorv001

Ok, thank you for the info...much appreciated.

vcoopers avatar Feb 24 '23 12:02 vcoopers

This error looks like it's related to the locally installed whisper transcription.

It has nothing to do with the TTS @connorv001

We need to try it out with other languages.

navopw avatar Feb 24 '23 13:02 navopw

Any ideas on how to fix the errors?

vcoopers avatar Feb 24 '23 15:02 vcoopers