File size: 22,755 Bytes
3ce989d
 
 
f5237ec
 
5bf19b3
 
5a007ca
3ce989d
4ea25cd
136ff40
0e508c8
048c3fc
04e2d2a
5a007ca
3ce989d
 
5a007ca
3ce989d
5bf19b3
1ed6720
04e2d2a
1ed6720
afdac46
fc05e1d
63ef86b
a6d4367
048c3fc
a6d4367
8047063
afdac46
2369fc5
ba3994f
 
 
09be04f
 
ba3994f
 
3ce989d
 
1ed6720
 
 
 
 
 
 
 
a4afe51
1ed6720
 
04e2d2a
1ed6720
 
 
 
 
 
 
 
04e2d2a
 
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a4afe51
1ed6720
 
 
 
04e2d2a
1ed6720
 
 
 
 
 
 
 
 
04e2d2a
1ed6720
 
 
 
 
 
 
 
 
04e2d2a
 
 
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
04e2d2a
 
 
 
 
 
1ed6720
 
 
04e2d2a
1ed6720
 
 
04e2d2a
1ed6720
 
 
 
 
 
04e2d2a
1ed6720
 
04e2d2a
 
 
 
 
 
 
 
 
 
 
 
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
04e2d2a
1ed6720
a4afe51
1ed6720
 
 
 
 
 
 
04e2d2a
1ed6720
 
 
 
 
 
 
 
 
 
 
04e2d2a
1ed6720
 
 
 
04e2d2a
 
 
1ed6720
 
 
 
 
 
 
 
04e2d2a
1ed6720
 
 
5bf19b3
 
1ed6720
 
 
f5237ec
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d1ed6b1
1ed6720
04e2d2a
1ed6720
 
 
 
 
 
 
 
04e2d2a
1ed6720
 
 
04e2d2a
 
 
 
1ed6720
 
 
 
 
04e2d2a
1ed6720
d1ed6b1
1ed6720
a4afe51
1ed6720
048c3fc
 
1ed6720
 
 
 
 
 
d1ed6b1
1ed6720
 
 
 
 
 
 
ee8b196
1ed6720
 
5bf19b3
 
 
1ed6720
1f53f73
a4afe51
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d1ed6b1
 
 
 
 
 
96154e7
 
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a4afe51
d1ed6b1
1ed6720
 
 
 
d1ed6b1
 
 
 
a4afe51
1ed6720
 
 
 
 
 
 
 
 
 
 
04e2d2a
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a4afe51
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a4afe51
1ed6720
 
 
 
 
 
 
 
 
 
a4afe51
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a4afe51
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
04e2d2a
1ed6720
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
"""
app.py

Gradio UI for interacting with the Anthropic API, Hume TTS API, and ElevenLabs TTS API.

Users enter a character description, which is processed using Claude by Anthropic to generate text.
The text is then synthesized into speech using different TTS provider APIs.
Users can compare the outputs and vote for their favorite in an interactive UI.
"""

# Standard Library Imports
import time
from concurrent.futures import ThreadPoolExecutor
from typing import Tuple

# Third-Party Library Imports
import gradio as gr

# Local Application Imports
from src import constants
from src.config import Config, logger
from src.custom_types import Option, OptionMap
from src.database.database import DBSessionMaker
from src.integrations import (
    AnthropicError,
    ElevenLabsError,
    HumeError,
    generate_text_with_claude,
    text_to_speech_with_elevenlabs,
    text_to_speech_with_hume,
)
from src.theme import CustomTheme
from src.utils import (
    choose_providers,
    create_shuffled_tts_options,
    determine_selected_option,
    submit_voting_results,
    validate_character_description_length,
)


class App:
    config: Config
    db_session_maker: DBSessionMaker

    def __init__(self, config: Config, db_session_maker: DBSessionMaker):
        self.config = config
        self.db_session_maker = db_session_maker

    def _generate_text(
        self,
        character_description: str,
    ) -> Tuple[dict, str]:
        """
        Validates the character_description and generates text using Anthropic API.

        Args:
            character_description (str): The user-provided text for character description.

        Returns:
            Tuple containing:
              - The generated text update (as a dict from gr.update).
              - The generated text string.

        Raises:
            gr.Error: On validation or API errors.
        """
        try:
            validate_character_description_length(character_description)
        except ValueError as ve:
            logger.warning(f"Validation error: {ve}")
            raise gr.Error(str(ve))

        try:
            generated_text = generate_text_with_claude(character_description, self.config)
            logger.info(f"Generated text ({len(generated_text)} characters).")
            return gr.update(value=generated_text), generated_text
        except AnthropicError as ae:
            logger.error(f"AnthropicError while generating text: {ae!s}")
            raise gr.Error(f'There was an issue communicating with the Anthropic API: "{ae.message}"')
        except Exception as e:
            logger.error(f"Unexpected error while generating text: {e}")
            raise gr.Error("Failed to generate text. Please try again later.")

    def _synthesize_speech(
        self,
        character_description: str,
        text: str,
        generated_text_state: str,
    ) -> Tuple[dict, dict, OptionMap, bool, str, str]:
        """
        Synthesizes two text-to-speech outputs, updates UI state components, and returns additional TTS metadata.

        This function generates TTS outputs using different providers based on the input text and its modification
        state. Depending on the selected providers, it may:
          - Synthesize one Hume and one ElevenLabs output (50% chance), or
          - Synthesize two Hume outputs (50% chance).

        The outputs are processed and shuffled, and the corresponding UI components for two audio players are updated.
        Additional metadata such as the comparison type, generation IDs, and state information are also returned.

        Args:
            character_description (str): The description of the character used for generating the voice.
            text (str): The text content to be synthesized into speech.
            generated_text_state (str): The previously generated text state, used to determine if the text has
                                        been modified.

        Returns:
            Tuple containing:
                - dict: Update for the first audio player (with autoplay enabled).
                - dict: Update for the second audio player.
                - OptionMap: A mapping of option constants to their corresponding TTS providers.
                - bool: Flag indicating whether the text was modified.
                - str: The original text that was synthesized.
                - str: The original character description.

        Raises:
            gr.Error: If any API or unexpected errors occur during the TTS synthesis process.
        """
        if not text:
            logger.warning("Skipping text-to-speech due to empty text.")
            raise gr.Error("Please generate or enter text to synthesize.")

        # Select 2 TTS providers based on whether the text has been modified.
        text_modified = text != generated_text_state
        provider_a, provider_b = choose_providers(text_modified, character_description)

        try:
            if provider_b == constants.HUME_AI:
                num_generations = 2
                # If generating 2 Hume outputs, do so in a single API call.
                result = text_to_speech_with_hume(character_description, text, num_generations, self.config)
                # Enforce that 4 values are returned.
                if not (isinstance(result, tuple) and len(result) == 4):
                    raise ValueError("Expected 4 values from Hume TTS call when generating 2 outputs")
                generation_id_a, audio_a, generation_id_b, audio_b = result
            else:
                with ThreadPoolExecutor(max_workers=2) as executor:
                    num_generations = 1
                    # Generate a single Hume output.
                    future_audio_a = executor.submit(
                        text_to_speech_with_hume, character_description, text, num_generations, self.config
                    )
                    # Generate a second TTS output from the second provider.
                    match provider_b:
                        case constants.ELEVENLABS:
                            future_audio_b = executor.submit(
                                text_to_speech_with_elevenlabs, character_description, text, self.config
                            )
                        case _:
                            # Additional TTS Providers can be added here.
                            raise ValueError(f"Unsupported provider: {provider_b}")

                    result_a = future_audio_a.result()
                    result_b = future_audio_b.result()
                    if isinstance(result_a, tuple) and len(result_a) >= 2:
                        generation_id_a, audio_a = result_a[0], result_a[1]
                    else:
                        raise ValueError("Unexpected return from text_to_speech_with_hume")
                    if isinstance(result_b, tuple) and len(result_b) >= 2:
                        generation_id_b, audio_b = result_b[0], result_b[1] # type: ignore
                    else:
                        raise ValueError("Unexpected return from text_to_speech_with_elevenlabs")

            # Shuffle options so that placement of options in the UI will always be random.
            option_a = Option(provider=provider_a, audio=audio_a, generation_id=generation_id_a)
            option_b = Option(provider=provider_b, audio=audio_b, generation_id=generation_id_b)
            options_map: OptionMap = create_shuffled_tts_options(option_a, option_b)

            option_a_audio = options_map["option_a"]["audio_file_path"]
            option_b_audio = options_map["option_b"]["audio_file_path"]

            return (
                gr.update(value=option_a_audio, visible=True, autoplay=True),
                gr.update(value=option_b_audio, visible=True),
                options_map,
                text_modified,
                text,
                character_description,
            )
        except ElevenLabsError as ee:
            logger.error(f"ElevenLabsError while synthesizing speech from text: {ee!s}")
            raise gr.Error(f'There was an issue communicating with the Elevenlabs API: "{ee.message}"')
        except HumeError as he:
            logger.error(f"HumeError while synthesizing speech from text: {he!s}")
            raise gr.Error(f'There was an issue communicating with the Hume API: "{he.message}"')
        except Exception as e:
            logger.error(f"Unexpected error during TTS generation: {e}")
            raise gr.Error("An unexpected error occurred. Please try again later.")

    def _vote(
        self,
        vote_submitted: bool,
        option_map: OptionMap,
        clicked_option_button: str,
        text_modified: bool,
        character_description: str,
        text: str,
    ) -> Tuple[bool, dict, dict, dict]:
        """
        Handles user voting.

        Args:
            vote_submitted (bool): True if a vote was already submitted.
            option_map (OptionMap): A dictionary mapping option labels to their details.
                Expected structure:
                {
                    'Option A': 'Hume AI',
                    'Option B': 'ElevenLabs',
                }
            clicked_option_button (str): The button that was clicked.

        Returns:
            A tuple of:
             - A boolean indicating if the vote was accepted.
             - A dict update for the selected vote button (showing provider and trophy emoji).
             - A dict update for the unselected vote button (showing provider).
             - A dict update for enabling vote interactions.
        """
        if not option_map or vote_submitted:
            return gr.skip(), gr.skip(), gr.skip(), gr.skip()

        selected_option, other_option = determine_selected_option(clicked_option_button)
        selected_provider = option_map[selected_option]["provider"]
        other_provider = option_map[other_option]["provider"]

        # Report voting results to be persisted to results DB.
        submit_voting_results(
            option_map,
            selected_option,
            text_modified,
            character_description,
            text,
            self.db_session_maker,
            self.config,
        )

        # Build button text, displaying the provider and voice name, appending the trophy emoji to the selected option.
        selected_label = f"{selected_provider} {constants.TROPHY_EMOJI}"
        other_label = f"{other_provider}"

        return (
            True,
            (
                gr.update(value=selected_label, variant="primary")
                if selected_option == constants.OPTION_A_KEY
                else gr.update(value=other_label, variant="secondary")
            ),
            (
                gr.update(value=other_label, variant="secondary")
                if selected_option == constants.OPTION_A_KEY
                else gr.update(value=selected_label, variant="primary")
            ),
            gr.update(interactive=True),
        )

    def _reset_ui(self) -> Tuple[dict, dict, dict, dict, OptionMap, bool]:
        """
        Resets UI state before generating new text.

        Returns:
            A tuple of updates for:
             - option_a_audio_player (clear audio)
             - option_b_audio_player (clear audio)
             - vote_button_a (disable and reset button text)
             - vote_button_b (disable and reset button text)
             - option_map_state (reset option map state)
             - vote_submitted_state (reset submitted vote state)
        """
        default_option_map: OptionMap = {
            "option_a": {"provider": constants.HUME_AI, "generation_id": None, "audio_file_path": ""},
            "option_b": {"provider": constants.HUME_AI, "generation_id": None, "audio_file_path": ""},
        }
        return (
            gr.update(value=None),
            gr.update(value=None, autoplay=False),
            gr.update(value=constants.SELECT_OPTION_A, variant="secondary"),
            gr.update(value=constants.SELECT_OPTION_B, variant="secondary"),
            default_option_map,  # Reset option_map_state as a default OptionMap
            False,
        )

    def _build_input_section(self) -> Tuple[gr.Dropdown, gr.Textbox, gr.Button]:
        """
        Builds the input section including the sample character description dropdown, character
        description input, and generate text button.
        """
        sample_character_description_dropdown = gr.Dropdown(
            choices=list(constants.SAMPLE_CHARACTER_DESCRIPTIONS.keys()),
            label="Choose a sample character description",
            value=None,
            interactive=True,
        )
        character_description_input = gr.Textbox(
            label="Character Description",
            placeholder="Enter a character description...",
            lines=3,
            max_lines=8,
            max_length=constants.CHARACTER_DESCRIPTION_MAX_LENGTH,
            show_copy_button=True,
        )
        generate_text_button = gr.Button("Generate Text", variant="secondary")
        return (
            sample_character_description_dropdown,
            character_description_input,
            generate_text_button,
        )

    def _build_output_section(self) -> Tuple[gr.Textbox, gr.Button, gr.Audio, gr.Audio, gr.Button, gr.Button]:
        """
        Builds the output section including text input, audio players, and vote buttons.
        """
        text_input = gr.Textbox(
            label="Input Text",
            placeholder="Enter or generate text for synthesis...",
            interactive=True,
            autoscroll=False,
            lines=3,
            max_lines=8,
            max_length=constants.CHARACTER_DESCRIPTION_MAX_LENGTH,
            show_copy_button=True,
        )
        synthesize_speech_button = gr.Button("Synthesize Speech", variant="primary")
        with gr.Row(equal_height=True):
            option_a_audio_player = gr.Audio(label=constants.OPTION_A_LABEL, type="filepath", interactive=False)
            option_b_audio_player = gr.Audio(label=constants.OPTION_B_LABEL, type="filepath", interactive=False)
        with gr.Row(equal_height=True):
            vote_button_a = gr.Button(constants.SELECT_OPTION_A, interactive=False)
            vote_button_b = gr.Button(constants.SELECT_OPTION_B, interactive=False)
        return (
            text_input,
            synthesize_speech_button,
            option_a_audio_player,
            option_b_audio_player,
            vote_button_a,
            vote_button_b,
        )

    def build_gradio_interface(self) -> gr.Blocks:
        """
        Builds and configures the Gradio user interface.

        Returns:
            gr.Blocks: The fully constructed Gradio UI layout.
        """
        custom_theme = CustomTheme()
        with gr.Blocks(
            title="Expressive TTS Arena",
            theme=custom_theme,
            fill_width=True,
            css_paths="src/assets/styles.css",
        ) as demo:
            # Title & instructions
            gr.Markdown("# Expressive TTS Arena")
            gr.Markdown(
                """
                1. **Choose or enter a character description**: Select a sample from the list or enter your own to guide
                text and voice generation.
                2. **Generate text**: Click **"Generate Text"** to create dialogue based on the character. The generated
                text will appear in the input field automatically—edit it if needed.
                3. **Synthesize speech**: Click **"Synthesize Speech"** to send your text and character description to
                two TTS APIs. Each API generates a voice and synthesizes speech in that voice.
                4. **Listen & compare**: Play both audio options and assess their expressiveness.
                5. **Vote for the best**: Click **"Select Option A"** or **"Select Option B"** to choose the most
                expressive output.
                """
            )

            # Build generate text section
            (
                sample_character_description_dropdown,
                character_description_input,
                generate_text_button,
            ) = self._build_input_section()

            # Build synthesize speech section
            (
                text_input,
                synthesize_speech_button,
                option_a_audio_player,
                option_b_audio_player,
                vote_button_a,
                vote_button_b,
            ) = self._build_output_section()

            # --- UI state components ---
            # Track character description used for text and voice generation
            character_description_state = gr.State("")
            # Track text used for speech synthesis
            text_state = gr.State("")
            # Track generated text state
            generated_text_state = gr.State("")
            # Track whether text that was used was generated or modified/custom
            text_modified_state = gr.State()
            # Track option map (option A and option B are randomized)
            option_map_state = gr.State({})  # OptionMap state as a dictionary
            # Track whether the user has voted for an option
            vote_submitted_state = gr.State(False)

            # --- Register event handlers ---

            # When a sample character description is chosen, update the character description textbox
            sample_character_description_dropdown.change(
                fn=lambda choice: constants.SAMPLE_CHARACTER_DESCRIPTIONS.get(choice, ""),
                inputs=[sample_character_description_dropdown],
                outputs=[character_description_input],
            )

            # Generate text button click handler chain:
            # 1. Disable the "Generate text" button
            # 2. Generate text
            # 3. Enable the "Generate text" button
            generate_text_button.click(
                fn=lambda: gr.update(interactive=False),
                inputs=[],
                outputs=[generate_text_button],
            ).then(
                fn=self._generate_text,
                inputs=[character_description_input],
                outputs=[text_input, generated_text_state],
            ).then(
                fn=lambda: gr.update(interactive=True),
                inputs=[],
                outputs=[generate_text_button],
            )

            # Synthesize speech button click event handler chain:
            # 1. Disable the "Synthesize speech" button
            # 2. Reset UI state
            # 3. Synthesize speech, load audio players, and display vote button
            # 4. Enable the "Synthesize speech" button and display vote buttons
            synthesize_speech_button.click(
                fn=lambda: (
                    gr.update(interactive=False),
                    gr.update(interactive=False),
                    gr.update(interactive=False),
                ),
                inputs=[],
                outputs=[synthesize_speech_button, vote_button_a, vote_button_b],
            ).then(
                fn=self._reset_ui,
                inputs=[],
                outputs=[
                    option_a_audio_player,
                    option_b_audio_player,
                    vote_button_a,
                    vote_button_b,
                    option_map_state,
                    vote_submitted_state,
                ],
            ).then(
                fn=self._synthesize_speech,
                inputs=[character_description_input, text_input, generated_text_state],
                outputs=[
                    option_a_audio_player,
                    option_b_audio_player,
                    option_map_state,
                    text_modified_state,
                    text_state,
                    character_description_state,
                ],
            ).then(
                fn=lambda: (
                    gr.update(interactive=True),
                    gr.update(interactive=True),
                    gr.update(interactive=True),
                ),
                inputs=[],
                outputs=[synthesize_speech_button, vote_button_a, vote_button_b],
            )

            # Vote button click event handlers
            vote_button_a.click(
                fn=lambda: (
                    gr.update(interactive=False),
                    gr.update(interactive=False),
                ),
                inputs=[],
                outputs=[vote_button_a, vote_button_b],
            ).then(
                fn=self._vote,
                inputs=[
                    vote_submitted_state,
                    option_map_state,
                    vote_button_a,
                    text_modified_state,
                    character_description_state,
                    text_state,
                ],
                outputs=[
                    vote_submitted_state,
                    vote_button_a,
                    vote_button_b,
                    synthesize_speech_button,
                ],
            )

            vote_button_b.click(
                fn=lambda: (
                    gr.update(interactive=False),
                    gr.update(interactive=False),
                ),
                inputs=[],
                outputs=[vote_button_a, vote_button_b],
            ).then(
                fn=self._vote,
                inputs=[
                    vote_submitted_state,
                    option_map_state,
                    vote_button_b,
                    text_modified_state,
                    character_description_state,
                    text_state,
                ],
                outputs=[
                    vote_submitted_state,
                    vote_button_a,
                    vote_button_b,
                    synthesize_speech_button,
                ],
            )

            # Reload audio player B with audio and set autoplay to True (workaround to play audio back-to-back)
            option_a_audio_player.stop(
                fn=lambda option_map: gr.update(
                    value=f"{option_map['option_b']['audio_file_path']}?t={int(time.time())}",
                    autoplay=True,
                ),
                inputs=[option_map_state],
                outputs=[option_b_audio_player],
            )

            # Enable voting after second audio option playback finishes
            option_b_audio_player.stop(
                fn=lambda _: gr.update(autoplay=False),
                inputs=[],
                outputs=[option_b_audio_player],
            )

        logger.debug("Gradio interface built successfully")
        return demo