From dc0dc6b3d2abc66971433ae606b70e7b9df5026d Mon Sep 17 00:00:00 2001
From: Raivis Dejus <raivisd@scandiweb.com>
Date: Sat, 13 Dec 2025 08:05:55 +0200
Subject: [PATCH] Adding speech extraction option to CLI (#1311)

---
 buzz/cli.py      | 6 ++++++
 docs/docs/cli.md | 3 ++-
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/buzz/cli.py b/buzz/cli.py
index b054e440..6fd56df0 100644
--- a/buzz/cli.py
+++ b/buzz/cli.py
@@ -102,6 +102,9 @@ def parse(app: Application, parser: QCommandLineParser):
         word_timestamp_option = QCommandLineOption(
             ["w", "word-timestamps"], "Generate word-level timestamps."
         )
+        extract_speech_option = QCommandLineOption(
+            ["e", "extract-speech"], "Extract speech from audio before transcribing."
+        )
         open_ai_access_token_option = QCommandLineOption(
             "openai-token",
             f"OpenAI access token. Use only when --model-type is {CommandLineModelType.OPEN_AI_WHISPER_API.value}. Defaults to your previously saved access token, if one exists.",
@@ -124,6 +127,7 @@ def parse(app: Application, parser: QCommandLineParser):
                 language_option,
                 initial_prompt_option,
                 word_timestamp_option,
+                extract_speech_option,
                 open_ai_access_token_option,
                 output_directory_option,
                 srt_option,
@@ -178,6 +182,7 @@ def parse(app: Application, parser: QCommandLineParser):
         initial_prompt = parser.value(initial_prompt_option)
 
         word_timestamps = parser.isSet(word_timestamp_option)
+        extract_speech = parser.isSet(extract_speech_option)
 
         output_formats: typing.Set[OutputFormat] = set()
         if parser.isSet(srt_option):
@@ -205,6 +210,7 @@ def parse(app: Application, parser: QCommandLineParser):
             language=language,
             initial_prompt=initial_prompt,
             word_level_timings=word_timestamps,
+            extract_speech=extract_speech,
             openai_access_token=openai_access_token,
         )
 
diff --git a/docs/docs/cli.md b/docs/docs/cli.md
index 751ed097..a8df135a 100644
--- a/docs/docs/cli.md
+++ b/docs/docs/cli.md
@@ -60,7 +60,8 @@ Options:
                                  (Yiddish), yo (Yoruba), zh (Chinese). Leave
                                  empty to detect language.
   -p, --prompt <prompt>          Initial prompt.
-  -w, --word-timestamps         Generate word-level timestamps. (available since 1.2.0)
+  -w, --word-timestamps          Generate word-level timestamps. (available since 1.2.0)
+  -e, --extract-speech           Extract speech from audio before transcribing. (available since 1.3.0)
   --openai-token <token>         OpenAI access token. Use only when
                                  --model-type is openaiapi. Defaults to your
                                  previously saved access token, if one exists.