Mohammad Hijjawi
1bca6ae4e4
The timestamps were built as str(timedelta(seconds=t))[:-3]. That only works when t has a fractional part: for a whole number of seconds str(timedelta) has no ".ffffff" suffix, so the slice cuts off the seconds instead, e.g. a segment starting at 0 s was written as "0:00 --> 0:00:01,500". Hours were also written with a single digit, while SRT expects HH:MM:SS,mmm. Format the timestamps explicitly as HH:MM:SS,mmm for both the .srt and the .txt output.
File description
-
./http_server.py It defines which files to server. Files are saved in ./web.
-
non_streaming_server.py WebSocket server for non-streaming models.
-
Orukeet Download and serve the multilingual Orukeet model using the existing NeMo transducer runtime. The same server options support Parakeet TDT v3.
-
vad-remove-non-speech-segments.py It uses silero-vad to remove non-speech segments and concatenate all speech segments into a single one.
-
vad-with-non-streaming-asr.py It shows how to use VAD with a non-streaming ASR model for speech recognition from a microphone
-
offline-speech-enhancement-gtcrn.py It shows how to use the offline speech denoiser API with GTCRN.
-
offline-speech-enhancement-dpdfnet.py It shows how to use the offline speech denoiser API with DPDFNet.
-
online-speech-enhancement-gtcrn.py It shows how to use the online speech denoiser API with GTCRN.
-
online-speech-enhancement-dpdfnet.py It shows how to use the online speech denoiser API with DPDFNet. Use 16 kHz DPDFNet models such as
dpdfnet_baseline.onnx,dpdfnet2.onnx,dpdfnet4.onnx, ordpdfnet8.onnxfor downstream ASR anddpdfnet2_8khz.onnxordpdfnet8_8khz.onnxfor 8 kHz enhancement. Usedpdfnet2_48khz_hr.onnxordpdfnet8_48khz_hr.onnxfor 48 kHz output. -
pocket-tts.py It shows how to use PocketTTS with the
GenerationConfigAPI. -
supertonic-tts.py It shows how to use SupertonicTTS with the
GenerationConfigAPI. -
zipvoice-tts.py It shows how to use ZipVoice for zero-shot TTS with the
GenerationConfigAPI. -
zipvoice-tts-play.py It shows how to use ZipVoice for zero-shot TTS and plays the generated audio while it is being synthesized.