tts: add speaker file support (ggml-org#12048)

* tts: add speaker file support

Signed-off-by: dm4 <[email protected]>

* tts: handle outetts-0.3

* tts : add new line in error message

---------

Signed-off-by: dm4 <[email protected]>
Co-authored-by: Georgi Gerganov <[email protected]>

Author: dm4

common/arg.cpp

@@ -2452,6 +2452,13 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
             params.vocoder.use_guide_tokens = true;
         }
     ).set_examples({LLAMA_EXAMPLE_TTS, LLAMA_EXAMPLE_SERVER}));
+    add_opt(common_arg(
+        {"--tts-speaker-file"}, "FNAME",
+        "speaker file path for audio generation",
+        [](common_params & params, const std::string & value) {
+            params.vocoder.speaker_file = value;
+        }
+    ).set_examples({LLAMA_EXAMPLE_TTS}));
 
     // model-specific
     add_opt(common_arg(

common/common.h

@@ -200,6 +200,8 @@ struct common_params_vocoder {
     std::string model     = ""; // model path                                                // NOLINT
     std::string model_url = ""; // model url to download                                     // NOLINT
 
+    std::string speaker_file = ""; // speaker file path                                      // NOLINT
+
     bool use_guide_tokens = false; // enable guide tokens to improve TTS accuracy            // NOLINT
 };