Update app.py

app.py

@@ -174,7 +174,7 @@ def predict(prompt, language, audio_file_pth, mic_file_path, use_mic, voice_clea
         t_latent=time.time()
         gpt_cond_latent, _, speaker_embedding = model.get_conditioning_latents(audio_path=speaker_wav)
         latent_calculation_time = time.time() - t_latent
-        metrics_text=f"Embedding calculation time: {latent_calculation_time:.2f} seconds\n"
+        ##metrics_text=f"Embedding calculation time: {latent_calculation_time:.2f} seconds\n"
         
         wav_chunks = []
 
@@ -190,7 +190,7 @@ def predict(prompt, language, audio_file_pth, mic_file_path, use_mic, voice_clea
             for i, chunk in enumerate(chunks):
                 if first_chunk:
                     first_chunk_time = time.time() - t_inference
-                    metrics_text+=f"Latency to first audio chunk: {round(first_chunk_time*1000)} seconds\n"
+                    metrics_text+=f"Latency to first audio chunk: {round(first_chunk_time*1000)} milliseconds\n"
                     first_chunk=False