From 59d0f0824461a0eafdc4c4fe64c6663ec05a36f3 Mon Sep 17 00:00:00 2001 From: mrq Date: Fri, 17 Feb 2023 06:06:50 +0000 Subject: [PATCH] https://arch.b4k.co/v/search/text/%22TAKE%20YOUR%20DAMN%20CLOTHES%20OFF%22/type/op/ --- src/utils.py | 2 +- src/webui.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/utils.py b/src/utils.py index 169959f..b8170ed 100755 --- a/src/utils.py +++ b/src/utils.py @@ -496,7 +496,7 @@ def prepare_dataset( files, outdir ): torchaudio.save(f"{outdir}/{sliced_name}", sliced_waveform, sampling_rate) - transcription.append(f"{sliced_name}|{segment['text'].trim()}") + transcription.append(f"{sliced_name}|{segment['text'].strip()}") idx = idx + 1 with open(f'{outdir}/whisper.json', 'w', encoding="utf-8") as f: diff --git a/src/webui.py b/src/webui.py index 976b292..0c0341e 100755 --- a/src/webui.py +++ b/src/webui.py @@ -428,7 +428,7 @@ def setup_gradio(): gr.Checkbox(label="Use CUDA for Voice Fixer", value=args.voice_fixer_use_cuda), gr.Checkbox(label="Force CPU for Conditioning Latents", value=args.force_cpu_for_conditioning_latents), gr.Textbox(label="Device Override", value=args.device_override), - gr.Dropdown(label="Whisper Model", value=args.whisper_model, choices=["tiny", "base", "small", "medium", "large"]), + gr.Dropdown(label="Whisper Model", value=args.whisper_model, choices=["tiny", "tiny.en", "base", "base.en", "small", "small.en", "medium", "medium.en", "large"]), ] gr.Button(value="Check for Updates").click(check_for_updates) gr.Button(value="Reload TTS").click(reload_tts)